Evidence · ANTs
antsRegistration
antsRegistration aligns one image with another through a sequence of stages, each a transform optimised against one or more metrics. LARMORX’s port is a replica of ANTs 2.6.5’s program on ITK 5.4.5, with every stage ANTs has (linear, SyN, B-spline SyN, B-spline, displacement-field, exponential and time-varying velocity-field) and every image and point-set metric. In reproduce it gives ANTs’ exit status, output and files in every compared case of its two records once ANTs uses correctly rounded maths (399 of 399 and 71 of 71). With ANTs as built, 355 of the 399 and 16 of the 71 match; the others differ only where glibc’s exp, log, sin or cos is not correctly rounded. On sMRIPrep’s own template registration it gives ANTs’ one-thread result with correctly rounded maths on any number of threads.
- Ports
- replica
larmorxApache-2.0: translated from ANTs 2.6.5's and ITK 5.4.5's source (Apache-2.0)
- replica
- Status
- replica validated, with its data panel (marked before the statuses' definition)
- Records
- standard comparison, full comparison (its velocity-field cases), the corrections' effects, data panel (its template registrations at full size), all against ANTs 2.6.5
- Reference
antsRegistration· API page ants-registration
What is compared
antsRegistration aligns a moving image with a fixed one through a sequence of stages, each a transform (rigid, affine, SyN, velocity fields and others) optimised against one or more metrics over a schedule of resolutions. sMRIPrep uses it for the brain extraction and the registration to the MNI template, fMRIPrep for the BOLD reference and the field map’s registration.
The replica runs beside ANTs 2.6.5’s antsRegistration on the same files with the same arguments. ANTs is the repository’s build (GCC 14.3), run as built and with correctly rounded maths, at one thread and at 2, 4 and 8 for its own variability.
larmorx ants antsRegistration -d 3 -o '[t1_to_mni_,t1_to_mni_Warped.nii.gz]' -t 'Rigid[0.05]' -m 'Mattes[mni.nii.gz,t1w.nii.gz,1,56,Regular,0.25]' -c '[100x100,1e-6,20]' -f 2x1 -s 2x1vox -t 'SyN[0.1,3,0]' -m 'CC[mni.nii.gz,t1w.nii.gz,1,4]' -c '[100x70x50x20,1e-6,10]' -f 8x4x2x1 -s 3x2x1x0voximport larmorx as lx
rigid = lx.ants.RegistrationStage(
transform="Rigid", gradient_step=0.1,
metrics=lx.ants.RegistrationMetric("Mattes", "fixed.nii.gz", "moving.nii.gz",
bins=32, sampling="Regular", sampling_percentage=0.25),
iterations=[1000, 500, 250], shrink_factors=[4, 2, 1], smoothing_sigmas=[2, 1, 0],
)
tvf = lx.ants.RegistrationStage(
"TimeVaryingVelocityField", 0.5, lx.ants.RegistrationMetric("CC", "fixed.nii.gz", "moving.nii.gz", radius=4),
iterations=[100, 70, 50, 20], shrink_factors=[8, 4, 2, 1], smoothing_sigmas=[3, 2, 1, 0],
number_of_time_indices=4, update_field_variance=3.0,
)
res = lx.ants.registration([rigid, tvf], warped=True, random_seed=42)The API page maps every option and stage. How to read what follows: how to read the evidence.
Its port
ANTs and ITK come under the Apache-2.0 licence, as LARMORX does, so the replica is in larmorx itself. It is translated from ANTs’ and ITK’s source, step by step, to give their results bit for bit. A clean room of ANTs’ programs is planned after FreeSurfer (roadmap).
The replica computes the C library’s maths (exp, log, sin, cos) correctly rounded, the same on every platform. Where glibc’s are not correctly rounded for a value a run meets, ANTs as built differs from the replica, and a registration’s iterations carry the change on.
The verdicts at a glance
antsRegistration 2.6.5 (standard comparison and data panel, 584 cases of the two records), by category and verdictSelect a segment to list its cases below.
In reproduce, every case that runs gives ANTs’ exit status, standard output with every iteration’s DIAGNOSTIC line, standard error and every file: the unmodified binary’s at one thread, or, where glibc’s exp, log, sin or cos is not correctly rounded for a value the run meets, the same binary’s with CORE-MATH’s correctly rounded functions preloaded. The other cases are inputs both programs refuse (93) and expected divergences (21): crashes of ANTs that the replica refuses with a message, and ANTs’ samplings seeded from the clock. No case is open.
In sMRIPrep: a real T1w registered to MNI space
fMRIPrep’s own registration of a T1w to MNI152NLin2009cAsym, run on every third voxel of its images as the standard comparison runs it (a full-size run takes ANTs about an hour at one thread; the full comparison and the data panel run it whole). Choose the output: the warped T1w, or one component of the displacement field that SyN found. Checkerboard and Swipe (under Compare) set the template against the warped T1w; the difference panels set the replica against ANTs as built next to ANTs against itself at 8 threads, on one scale.
Worked example
sMRIPrep's registration of a real T1w to MNI space
What this shows: In reproduce, on 1 and 12 threads, the replica gives ANTs’ one-thread result with correctly rounded maths bit for bit: the warped T1w, every value of the displacement field and all 865 lines of its output (addresses, times and paths masked, as the record compares them). ANTs as built differs only where glibc’s exp, log, sin or cos is not correctly rounded, and the optimiser’s iterations carry the change: up to 117 in the warped T1w, and a voxel’s displacement moves by up to 1.08 mm in the brain. ANTs’ own thread counts move its result further: up to 240 and 1.49 mm, 4 threads against one. No correction by default changes this call.
- Replica against ANTs antsRegistration 2.6.5Identical with correctly rounded maths
fMRIPrep 25.2.6’s template registration of OpenNeuro ds000005 sub-01 (its own arguments, from the fMRIPrep run’s command.txt) on every third voxel of its images, as the record’s standard comparison runs it, given to ANTs 2.6.5 and to the replica. Choose the output (the warped T1w or a component of the displacement field), the plane and the panels; Checkerboard and Swipe compare the template with the warped T1w.
- Data
- OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01’s T1w as fMRIPrep preprocessed it (PDDL-1.0), × its brain mask; the template, MNI ICBM 152 nonlinear asymmetric 2009c, © McConnell Brain Imaging Centre, Montreal Neurological Institute (notice)
- Stages
- Rigid, Affine and SyN, exactly fMRIPrep’s
- Inputs
- as the record’s case runs it: the masked template and T1w nipype did not keep (
fixed_masked,moving_masked) remade from the run’s outputs, the initial transform it did not keep (transform.mat) replaced by[fixed,moving,1](centres of mass), every image on every third voxel - Command
antsRegistration --collapse-output-transforms 1 --dimensionality 3 --float 1 --initial-moving-transform [transform.mat,0] --initialize-transforms-per-stage 0 --interpolation LanczosWindowedSinc --output [ants_t1_to_mni,ants_t1_to_mni_Warped.nii.gz] --transform Rigid[0.05] --metric Mattes[fixed_masked.nii.gz,moving_masked.nii.gz,1,56,Regular,0.25] --convergence [100x100,1e-06,20] --smoothing-sigmas 2.0x1.0vox --shrink-factors 2x1 --use-histogram-matching 1 --transform Affine[0.08] --metric Mattes[fixed_masked.nii.gz,moving_masked.nii.gz,1,56,Regular,0.25] --convergence [100x100,1e-06,20] --smoothing-sigmas 1.0x0.0vox --shrink-factors 2x1 --use-histogram-matching 1 --transform SyN[0.1,3.0,0.0] --metric CC[fixed_masked.nii.gz,moving_masked.nii.gz,1,4,None,1] --convergence [100x70x50x20,1e-06,10] --smoothing-sigmas 3.0x2.0x1.0x0.0vox --shrink-factors 8x4x2x1 --use-histogram-matching 1 --winsorize-image-intensities [0.005,0.995] --write-composite-transform 1- Programs
- ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0,
ANTS_RANDOM_SEED=42; the replica at commit1dfe56c0dc60 - Output
- standard output (every iteration’s DIAGNOSTIC line, normalised as the record does): 865 lines; identical to ANTs with CORE-MATH’s, 388 differ from ANTs as built
- Grid
- 65 × 77 × 65 voxels of 3 × 3 × 3 mm; shown: 55 × 67 × 57 around the brain
Data shown
- Mixed-gambles task (OpenNeuro ds000005, version 00001), Open Data Commons Public Domain Dedication and Licence 1.0.
- MNI152NLin2009cAsym template (TemplateFlow, version tpl-MNI152NLin2009cAsym git 15d7c02160f79f5218d2545b4febebeecc11531d), the MNI ICBM 152 copyright notice.






Bit-identical325,325 of 325,325 voxels equal
Axial slice 28 of 57; the crosshair where the replica and ANTs as built differ most in it; the warped T1w in the brain's 0.5–99.5 % window; the differences on one symmetric log scale over six decades below the largest, any non-zero difference tinted. The interactive viewer needs JavaScript; the metrics below are complete without it.
What separates ANTs' result at fMRIPrep's 8 threads from the replica's
- ANTs antsRegistration 2.6.5, 8 threadsas built (fMRIPrep ran 8)
- 1 thread instead of 8max |Δ| 128
- ANTs antsRegistration 2.6.5, 1 thread, as builtthe repository's build (GCC 14.3), glibc's maths
- correctly rounded maths instead of glibc'smax |Δ| 117
- ANTs antsRegistration 2.6.5, correctly rounded maths, 1 threadthe same binary, CORE-MATH preloaded
- replica1 and 12 threads, default and reproduce
Each box holds runs that give the same bytes, every output (the warped T1w and the three components of the displacement field); the largest |Δ| of a link is over all four. ANTs’ thread count and glibc’s maths are the only two causes, and the replica computes ANTs’ one-thread result with correctly rounded maths, on any number of threads.
Every comparison, computed in float64 over every voxel of the outputs
| Comparison (A − B), verdict and reason | Output | max |Δ| | Values that differ | RMS of Δ | max |Δ| ÷ range | 1 − NCC |
|---|---|---|---|---|---|---|
| The original against itself: the yardstick | ||||||
| ANTs antsRegistration 2.6.5, 8 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 128 | 124,445 of 325,325 (38.3 %) | 6.17 | 0.0659 | 0.000206 |
| Field x (LPS) | 1.25 | 276,814 of 325,325 (85.1 %) | 0.151 | 0.108 | 0.0121 | |
| Field y (LPS) | 0.809 | 276,519 of 325,325 (85 %) | 0.152 | 0.0674 | 0.005 | |
| Field z (LPS) | 0.964 | 276,232 of 325,325 (84.9 %) | 0.121 | 0.076 | 0.0031 | |
| ANTs antsRegistration 2.6.5 with correctly rounded maths − as built, 1 threadreferenceThe same ANTs binary, its glibc functions replaced by correctly rounded ones. platform maths | Warped T1w | 117 | 124,089 of 325,325 (38.1 %) | 6.27 | 0.0601 | 0.000213 |
| Field x (LPS) | 0.759 | 276,747 of 325,325 (85.1 %) | 0.0997 | 0.0659 | 0.00611 | |
| Field y (LPS) | 0.887 | 276,387 of 325,325 (85 %) | 0.116 | 0.0738 | 0.00277 | |
| Field z (LPS) | 0.898 | 276,137 of 325,325 (84.9 %) | 0.103 | 0.0708 | 0.00232 | |
| ANTs antsRegistration 2.6.5, 2 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 102 | 123,992 of 325,325 (38.1 %) | 5.52 | 0.0526 | 0.000165 |
| Field x (LPS) | 0.905 | 276,750 of 325,325 (85.1 %) | 0.1 | 0.0786 | 0.00552 | |
| Field y (LPS) | 0.546 | 276,401 of 325,325 (85 %) | 0.104 | 0.0455 | 0.0027 | |
| Field z (LPS) | 0.889 | 276,124 of 325,325 (84.9 %) | 0.0877 | 0.0701 | 0.00144 | |
| ANTs antsRegistration 2.6.5, 4 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 240 | 124,961 of 325,325 (38.4 %) | 9.15 | 0.123 | 0.000454 |
| Field x (LPS) | 1.22 | 276,809 of 325,325 (85.1 %) | 0.189 | 0.106 | 0.0156 | |
| Field y (LPS) | 1.1 | 276,370 of 325,325 (85 %) | 0.192 | 0.0918 | 0.00921 | |
| Field z (LPS) | 1.37 | 276,285 of 325,325 (84.9 %) | 0.202 | 0.108 | 0.00801 | |
| The replica against the original | ||||||
| replica − ANTs antsRegistration 2.6.5, 1 thread, as builtIdentical with correctly rounded mathsOnly where glibc’s exp, log, sin or cos is not correctly rounded: with CORE-MATH preloaded, the same ANTs binary gives the replica’s output bit for bit (the row against correctly rounded maths). The change then grows through the optimiser’s iterations, as ANTs’ own threads do. platform maths | Warped T1w | 117 | 124,089 of 325,325 (38.1 %) | 6.27 | 0.0601 | 0.000213 |
| Field x (LPS) | 0.759 | 276,747 of 325,325 (85.1 %) | 0.0997 | 0.0659 | 0.00611 | |
| Field y (LPS) | 0.887 | 276,387 of 325,325 (85 %) | 0.116 | 0.0738 | 0.00277 | |
| Field z (LPS) | 0.898 | 276,137 of 325,325 (84.9 %) | 0.103 | 0.0708 | 0.00232 | |
| replica − ANTs antsRegistration 2.6.5, 1 thread, correctly rounded mathsBit-identicalWith correctly rounded maths, ANTs computes exactly what the replica computes: the warped image and every value of the displacement field. | every output | identical: 0 of 325,325 values differ | ||||
| The replica against itself: thread counts, fix and reproduce | ||||||
| replica, 12 threads − 1 threadBit-identicalThe replica gives ANTs’ single-thread result on any number of threads. | every output | identical: 0 of 325,325 values differ | ||||
| replica, fix − reproduceBit-identicalNo correction of the register applies to this command: both modes give the same bytes. | every output | identical: 0 of 325,325 values differ | ||||
| replica, fix, 12 threads − 1 threadBit-identicalThe same bytes on any number of threads. | every output | identical: 0 of 325,325 values differ | ||||
Δ is A − B. Every number is computed in float64 over all 325,325 voxels of the 65 × 77 × 65 outputs (the field: each component), not only the brain shown; the range is B’s maximum minus its minimum. The rows the original against itself are ANTs’ own variability on this input. The field’s components are in mm, in ITK’s physical space (LPS).
The linear stages’ result: the affine of the composite transform (Rigid then Affine, collapsed; AffineTransform_float_3_3)
| Parameter | ANTs, 1 thread, as built | ANTs, 1 thread, CORE-MATH | replica | replica − ANTs as built | replica − ANTs with CORE-MATH | ANTs 8 threads − 1 thread |
|---|---|---|---|---|---|---|
| matrix 1,1 | 0.9440594 | 0.9426623 | 0.9426623 | −0.0014 | 0 | −0.00384 |
| matrix 1,2 | 9.248753e−4 | −2.785155e−4 | −2.785155e−4 | −0.0012 | 0 | −0.00257 |
| matrix 1,3 | 0.0035838322 | 0.0042304373 | 0.0042304373 | 0.000647 | 0 | −0.00216 |
| matrix 2,1 | 0.007971352 | 0.008520667 | 0.008520667 | 0.000549 | 0 | 0.00015 |
| matrix 2,2 | 0.8210773 | 0.82226926 | 0.82226926 | 0.00119 | 0 | −0.000255 |
| matrix 2,3 | 0.2969963 | 0.29518113 | 0.29518113 | −0.00182 | 0 | 0.00265 |
| matrix 3,1 | −0.013978851 | −0.014700439 | −0.014700439 | −0.000722 | 0 | −0.000943 |
| matrix 3,2 | −0.32897782 | −0.32976612 | −0.32976612 | −0.000788 | 0 | −0.000327 |
| matrix 3,3 | 0.75971 | 0.757824 | 0.757824 | −0.00189 | 0 | −0.00471 |
| translation x | 1.9234301 | 1.9703183 | 1.9703183 | 0.0469 | 0 | 0.0726 |
| translation y | 11.570224 | 11.559067 | 11.559067 | −0.0112 | 0 | 0.0814 |
| translation z | −17.417015 | −17.481945 | −17.481945 | −0.0649 | 0 | −0.00817 |
The parameters as the files store them (float32: antsRegistration ran with --float 1), printed with the digits that identify each value; the matrix row by row in ITK’s physical space (LPS), the translation in mm. The centre of rotation is the same in every run (0.250445, 21.9911, 9.73538 mm).
The displacement fields: how far a voxel’s displacement moves (the length of the difference vector, mm), over the template’s brain
| Comparison | largest (mm) | 99th percentile (mm) | RMS (mm) | voxels that differ, of 69,880 |
|---|---|---|---|---|
| replica − ANTs as built, 1 thread | 1.08 | 0.63 | 0.296 | 69,880 |
| replica − ANTs with CORE-MATH, 1 thread | 0 | 0 | 0 | 0 |
| ANTs 2 threads − 1 thread | 1.04 | 0.61 | 0.267 | 69,880 |
| ANTs 4 threads − 1 thread | 1.49 | 1.08 | 0.568 | 69,880 |
| ANTs 8 threads − 1 thread | 1.25 | 0.846 | 0.403 | 69,880 |
| ANTs with CORE-MATH − as built, 1 thread | 1.08 | 0.63 | 0.296 | 69,880 |
ANTs’ own runs at 2, 4 and 8 threads are the yardstick: the replica’s difference from ANTs as built is the size of a change of ANTs’ thread count, and nothing with correctly rounded maths.
Provenance of this example
- Data: OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01, PDDL-1.0; MNI ICBM 152 nonlinear asymmetric 2009c (Fonov et al., 2009, 2011), from TemplateFlow, the MNI ICBM 152 notice (static/hero/LICENSE-MNI.txt)
- Inputs: fixed (sha256 b06d202af1954dbf…); moving (sha256 d358573fa558039d…)
- Runs: figures cache entry
ants-registration/fmriprep-ds000005-template-syn-small, computed 2026-10-21T15:00:09+00:00 with larmorx1dfe56c0dc60; ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0; every command, its environment, threads and exit status are in itsRUN_INFO.json - Stage under test: SyN[0.1,3,0]
Every number above is in ants-registration.json, exported by python -m larmorx_validation evidence ants-registration.
Worked example
The same registration with a B-spline SyN stage
What this shows: In reproduce, on 1 and 12 threads, the replica gives ANTs’ one-thread result with correctly rounded maths bit for bit: the warped T1w, the B-spline smoothed field and all 857 lines of its output (addresses, times and paths masked, as the record compares them). ANTs as built differs only where glibc’s maths is not correctly rounded: up to 57 in the warped T1w, and a voxel’s displacement moves by up to 0.51 mm in the brain. ANTs’ own thread counts move its result further: up to 232 at 2 threads and 1.05 mm at 4 threads, against one. No correction by default changes this call.
- Replica against ANTs antsRegistration 2.6.5Identical with correctly rounded maths
The command above with its SyN stage replaced by BSplineSyN[0.1,26,0,3], as antsRegistrationSyN.sh’s -t b runs it (the record’s real-bsplinesyn-small case): the update field and the total field are B-spline smoothed.
- Data
- OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01’s T1w as fMRIPrep preprocessed it (PDDL-1.0), × its brain mask; the template, MNI ICBM 152 nonlinear asymmetric 2009c, © McConnell Brain Imaging Centre, Montreal Neurological Institute (notice)
- Stages
- Rigid and Affine as fMRIPrep’s, then antsRegistrationSyN.sh’s BSplineSyN[0.1,26,0,3] (its
-t b) in place of SyN[0.1,3,0] - Inputs
- as the record’s case runs it: the masked template and T1w nipype did not keep (
fixed_masked,moving_masked) remade from the run’s outputs, the initial transform it did not keep (transform.mat) replaced by[fixed,moving,1](centres of mass), every image on every third voxel - Command
antsRegistration --collapse-output-transforms 1 --dimensionality 3 --float 1 --initial-moving-transform [transform.mat,0] --initialize-transforms-per-stage 0 --interpolation LanczosWindowedSinc --output [ants_t1_to_mni,ants_t1_to_mni_Warped.nii.gz] --transform Rigid[0.05] --metric Mattes[fixed_masked.nii.gz,moving_masked.nii.gz,1,56,Regular,0.25] --convergence [100x100,1e-06,20] --smoothing-sigmas 2.0x1.0vox --shrink-factors 2x1 --use-histogram-matching 1 --transform Affine[0.08] --metric Mattes[fixed_masked.nii.gz,moving_masked.nii.gz,1,56,Regular,0.25] --convergence [100x100,1e-06,20] --smoothing-sigmas 1.0x0.0vox --shrink-factors 2x1 --use-histogram-matching 1 --transform BSplineSyN[0.1,26,0,3] --metric CC[fixed_masked.nii.gz,moving_masked.nii.gz,1,4,None,1] --convergence [100x70x50x20,1e-06,10] --smoothing-sigmas 3.0x2.0x1.0x0.0vox --shrink-factors 8x4x2x1 --use-histogram-matching 1 --winsorize-image-intensities [0.005,0.995] --write-composite-transform 1- Programs
- ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0,
ANTS_RANDOM_SEED=42; the replica at commit1dfe56c0dc60 - Output
- standard output (every iteration’s DIAGNOSTIC line, normalised as the record does): 857 lines; identical to ANTs with CORE-MATH’s, 380 differ from ANTs as built
- Grid
- 65 × 77 × 65 voxels of 3 × 3 × 3 mm; shown: 55 × 67 × 57 around the brain
Data shown
- Mixed-gambles task (OpenNeuro ds000005, version 00001), Open Data Commons Public Domain Dedication and Licence 1.0.
- MNI152NLin2009cAsym template (TemplateFlow, version tpl-MNI152NLin2009cAsym git 15d7c02160f79f5218d2545b4febebeecc11531d), the MNI ICBM 152 copyright notice.






Bit-identical325,325 of 325,325 voxels equal
Axial slice 28 of 57; the crosshair where the replica and ANTs as built differ most in it; the warped T1w in the brain's 0.5–99.5 % window; the differences on one symmetric log scale over six decades below the largest, any non-zero difference tinted. The interactive viewer needs JavaScript; the metrics below are complete without it.
What separates ANTs' result at fMRIPrep's 8 threads from the replica's
- ANTs antsRegistration 2.6.5, 8 threadsas built (fMRIPrep ran 8)
- 1 thread instead of 8max |Δ| 201
- ANTs antsRegistration 2.6.5, 1 thread, as builtthe repository's build (GCC 14.3), glibc's maths
- correctly rounded maths instead of glibc'smax |Δ| 56.6
- ANTs antsRegistration 2.6.5, correctly rounded maths, 1 threadthe same binary, CORE-MATH preloaded
- replica1 and 12 threads, default and reproduce
Each box holds runs that give the same bytes, every output (the warped T1w and the three components of the displacement field); the largest |Δ| of a link is over all four. ANTs’ thread count and glibc’s maths are the only two causes, and the replica computes ANTs’ one-thread result with correctly rounded maths, on any number of threads.
Every comparison, computed in float64 over every voxel of the outputs
| Comparison (A − B), verdict and reason | Output | max |Δ| | Values that differ | RMS of Δ | max |Δ| ÷ range | 1 − NCC |
|---|---|---|---|---|---|---|
| The original against itself: the yardstick | ||||||
| ANTs antsRegistration 2.6.5, 8 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 201 | 128,589 of 325,325 (39.5 %) | 9.23 | 0.0932 | 0.000459 |
| Field x (LPS) | 0.775 | 272,063 of 325,325 (83.6 %) | 0.0812 | 0.13 | 0.00978 | |
| Field y (LPS) | 0.501 | 272,065 of 325,325 (83.6 %) | 0.0624 | 0.0703 | 0.00367 | |
| Field z (LPS) | 0.975 | 272,154 of 325,325 (83.7 %) | 0.0875 | 0.097 | 0.0049 | |
| ANTs antsRegistration 2.6.5 with correctly rounded maths − as built, 1 threadreferenceThe same ANTs binary, its glibc functions replaced by correctly rounded ones. platform maths | Warped T1w | 56.6 | 127,867 of 325,325 (39.3 %) | 2.76 | 0.0263 | 4.1e−5 |
| Field x (LPS) | 0.51 | 272,052 of 325,325 (83.6 %) | 0.0395 | 0.0854 | 0.00262 | |
| Field y (LPS) | 0.289 | 272,056 of 325,325 (83.6 %) | 0.0284 | 0.0406 | 0.000789 | |
| Field z (LPS) | 0.368 | 272,148 of 325,325 (83.7 %) | 0.0367 | 0.0367 | 0.000889 | |
| ANTs antsRegistration 2.6.5, 2 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 232 | 128,726 of 325,325 (39.6 %) | 11.9 | 0.108 | 0.000767 |
| Field x (LPS) | 0.648 | 272,061 of 325,325 (83.6 %) | 0.106 | 0.109 | 0.0197 | |
| Field y (LPS) | 0.915 | 272,064 of 325,325 (83.6 %) | 0.0892 | 0.128 | 0.00784 | |
| Field z (LPS) | 0.783 | 272,150 of 325,325 (83.7 %) | 0.0924 | 0.0779 | 0.00551 | |
| ANTs antsRegistration 2.6.5, 4 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 175 | 127,953 of 325,325 (39.3 %) | 7.6 | 0.0813 | 0.000312 |
| Field x (LPS) | 0.608 | 272,059 of 325,325 (83.6 %) | 0.0628 | 0.102 | 0.00673 | |
| Field y (LPS) | 0.618 | 272,060 of 325,325 (83.6 %) | 0.0705 | 0.0868 | 0.00514 | |
| Field z (LPS) | 0.732 | 272,150 of 325,325 (83.7 %) | 0.0495 | 0.0728 | 0.00165 | |
| The replica against the original | ||||||
| replica − ANTs antsRegistration 2.6.5, 1 thread, as builtIdentical with correctly rounded mathsOnly where glibc’s exp, log, sin or cos is not correctly rounded: with CORE-MATH preloaded, the same ANTs binary gives the replica’s output bit for bit (the row against correctly rounded maths). The change then grows through the optimiser’s iterations, as ANTs’ own threads do. platform maths | Warped T1w | 56.6 | 127,867 of 325,325 (39.3 %) | 2.76 | 0.0263 | 4.1e−5 |
| Field x (LPS) | 0.51 | 272,052 of 325,325 (83.6 %) | 0.0395 | 0.0854 | 0.00262 | |
| Field y (LPS) | 0.289 | 272,056 of 325,325 (83.6 %) | 0.0284 | 0.0406 | 0.000789 | |
| Field z (LPS) | 0.368 | 272,148 of 325,325 (83.7 %) | 0.0367 | 0.0367 | 0.000889 | |
| replica − ANTs antsRegistration 2.6.5, 1 thread, correctly rounded mathsBit-identicalWith correctly rounded maths, ANTs computes exactly what the replica computes: the warped image and every value of the displacement field. | every output | identical: 0 of 325,325 values differ | ||||
| The replica against itself: thread counts, fix and reproduce | ||||||
| replica, 12 threads − 1 threadBit-identicalThe replica gives ANTs’ single-thread result on any number of threads. | every output | identical: 0 of 325,325 values differ | ||||
| replica, fix − reproduceBit-identicalNo correction of the register applies to this command: both modes give the same bytes. | every output | identical: 0 of 325,325 values differ | ||||
| replica, fix, 12 threads − 1 threadBit-identicalThe same bytes on any number of threads. | every output | identical: 0 of 325,325 values differ | ||||
Δ is A − B. Every number is computed in float64 over all 325,325 voxels of the 65 × 77 × 65 outputs (the field: each component), not only the brain shown; the range is B’s maximum minus its minimum. The rows the original against itself are ANTs’ own variability on this input. The field’s components are in mm, in ITK’s physical space (LPS).
The linear stages’ result: the affine of the composite transform (Rigid then Affine, collapsed; AffineTransform_float_3_3)
| Parameter | ANTs, 1 thread, as built | ANTs, 1 thread, CORE-MATH | replica | replica − ANTs as built | replica − ANTs with CORE-MATH | ANTs 8 threads − 1 thread |
|---|---|---|---|---|---|---|
| matrix 1,1 | 0.9440594 | 0.9426623 | 0.9426623 | −0.0014 | 0 | −0.00413 |
| matrix 1,2 | 9.248753e−4 | −2.785155e−4 | −2.785155e−4 | −0.0012 | 0 | 0.000144 |
| matrix 1,3 | 0.0035838322 | 0.0042304373 | 0.0042304373 | 0.000647 | 0 | 0.00139 |
| matrix 2,1 | 0.007971352 | 0.008520667 | 0.008520667 | 0.000549 | 0 | −0.00191 |
| matrix 2,2 | 0.8210773 | 0.82226926 | 0.82226926 | 0.00119 | 0 | −0.00145 |
| matrix 2,3 | 0.2969963 | 0.29518113 | 0.29518113 | −0.00182 | 0 | 0.00557 |
| matrix 3,1 | −0.013978851 | −0.014700439 | −0.014700439 | −0.000722 | 0 | −0.00352 |
| matrix 3,2 | −0.32897782 | −0.32976612 | −0.32976612 | −0.000788 | 0 | −0.00332 |
| matrix 3,3 | 0.75971 | 0.757824 | 0.757824 | −0.00189 | 0 | −0.00695 |
| translation x | 1.9234301 | 1.9703183 | 1.9703183 | 0.0469 | 0 | 0.012 |
| translation y | 11.570224 | 11.559067 | 11.559067 | −0.0112 | 0 | 0.112 |
| translation z | −17.417015 | −17.481945 | −17.481945 | −0.0649 | 0 | −0.104 |
The parameters as the files store them (float32: antsRegistration ran with --float 1), printed with the digits that identify each value; the matrix row by row in ITK’s physical space (LPS), the translation in mm. The centre of rotation is the same in every run (0.250445, 21.9911, 9.73538 mm).
The displacement fields: how far a voxel’s displacement moves (the length of the difference vector, mm), over the template’s brain
| Comparison | largest (mm) | 99th percentile (mm) | RMS (mm) | voxels that differ, of 69,880 |
|---|---|---|---|---|
| replica − ANTs as built, 1 thread | 0.511 | 0.217 | 0.105 | 69,880 |
| replica − ANTs with CORE-MATH, 1 thread | 0 | 0 | 0 | 0 |
| ANTs 2 threads − 1 thread | 0.901 | 0.574 | 0.278 | 69,880 |
| ANTs 4 threads − 1 thread | 1.05 | 0.402 | 0.17 | 69,880 |
| ANTs 8 threads − 1 thread | 0.988 | 0.496 | 0.223 | 69,880 |
| ANTs with CORE-MATH − as built, 1 thread | 0.511 | 0.217 | 0.105 | 69,880 |
ANTs’ own runs at 2, 4 and 8 threads are the yardstick: the replica’s difference from ANTs as built is the size of a change of ANTs’ thread count, and nothing with correctly rounded maths.
Provenance of this example
- Data: OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01, PDDL-1.0; MNI ICBM 152 nonlinear asymmetric 2009c (Fonov et al., 2009, 2011), from TemplateFlow, the MNI ICBM 152 notice (static/hero/LICENSE-MNI.txt)
- Inputs: fixed (sha256 b06d202af1954dbf…); moving (sha256 d358573fa558039d…)
- Runs: figures cache entry
ants-registration/fmriprep-ds000005-template-bsplinesyn-small, computed 2026-10-21T15:31:20+00:00 with larmorx1dfe56c0dc60; ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0; every command, its environment, threads and exit status are in itsRUN_INFO.json - Stage under test: BSplineSyN[0.1,26,0,3]
Every number above is in ants-registration.json, exported by python -m larmorx_validation evidence ants-registration.
A velocity-field registration
The same command with its SyN stage replaced by a time-varying velocity field: the deformation is the flow of a velocity field over space and four time points, integrated from time 0 to 1 by fourth-order Runge–Kutta, and its inverse the flow back. ANTs has no script that sets this stage, so it takes SyN’s levels, iterations, metric and smoothing. The exponential stages (a stationary velocity field, its displacement the exponential by scaling and squaring) and the time-varying B-spline field are in the record the same way, on the same images.
Worked example
The same registration with a time-varying velocity field
What this shows: In reproduce, on 1 and 12 threads, the replica gives ANTs’ one-thread result with correctly rounded maths bit for bit: the warped T1w, the field integrated from the velocity field and all 988 lines of its output (addresses, times and paths masked, as the record compares them). ANTs as built differs only where glibc’s maths is not correctly rounded: up to 53 in the warped T1w, and a voxel’s displacement moves by up to 0.27 mm in the brain. ANTs’ own thread counts move its result further: up to 283 and 0.92 mm, 4 threads against one. No correction by default changes this call.
- Replica against ANTs antsRegistration 2.6.5Identical with correctly rounded maths
The command above with its SyN stage replaced by TimeVaryingVelocityField[0.5,4,3,0,0,0] (the record’s real-time-varying-small case): a velocity field over space and four time points, its displacement the flow from time 0 to 1 integrated by fourth-order Runge–Kutta, the update smoothed in space with SyN’s variance of 3 voxels².
- Data
- OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01’s T1w as fMRIPrep preprocessed it (PDDL-1.0), × its brain mask; the template, MNI ICBM 152 nonlinear asymmetric 2009c, © McConnell Brain Imaging Centre, Montreal Neurological Institute (notice)
- Stages
- Rigid and Affine as fMRIPrep’s, then TimeVaryingVelocityField[0.5,4,3,0,0,0] in place of SyN[0.1,3,0], with SyN’s levels, iterations and CC metric
- Inputs
- as the record’s case runs it: the masked template and T1w nipype did not keep (
fixed_masked,moving_masked) remade from the run’s outputs, the initial transform it did not keep (transform.mat) replaced by[fixed,moving,1](centres of mass), every image on every third voxel - Command
antsRegistration --collapse-output-transforms 1 --dimensionality 3 --float 1 --initial-moving-transform [transform.mat,0] --initialize-transforms-per-stage 0 --interpolation LanczosWindowedSinc --output [ants_t1_to_mni,ants_t1_to_mni_Warped.nii.gz] --transform Rigid[0.05] --metric Mattes[fixed_masked.nii.gz,moving_masked.nii.gz,1,56,Regular,0.25] --convergence [100x100,1e-06,20] --smoothing-sigmas 2.0x1.0vox --shrink-factors 2x1 --use-histogram-matching 1 --transform Affine[0.08] --metric Mattes[fixed_masked.nii.gz,moving_masked.nii.gz,1,56,Regular,0.25] --convergence [100x100,1e-06,20] --smoothing-sigmas 1.0x0.0vox --shrink-factors 2x1 --use-histogram-matching 1 --transform TimeVaryingVelocityField[0.5,4,3,0,0,0] --metric CC[fixed_masked.nii.gz,moving_masked.nii.gz,1,4,None,1] --convergence [100x70x50x20,1e-06,10] --smoothing-sigmas 3.0x2.0x1.0x0.0vox --shrink-factors 8x4x2x1 --use-histogram-matching 1 --winsorize-image-intensities [0.005,0.995] --write-composite-transform 1- Programs
- ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0,
ANTS_RANDOM_SEED=42; the replica at commita9401a69f759 - Output
- standard output (every iteration’s DIAGNOSTIC line, normalised as the record does): 988 lines; identical to ANTs with CORE-MATH’s, 511 differ from ANTs as built
- Grid
- 65 × 77 × 65 voxels of 3 × 3 × 3 mm; shown: 55 × 67 × 57 around the brain
Data shown
- Mixed-gambles task (OpenNeuro ds000005, version 00001), Open Data Commons Public Domain Dedication and Licence 1.0.
- MNI152NLin2009cAsym template (TemplateFlow, version tpl-MNI152NLin2009cAsym git 15d7c02160f79f5218d2545b4febebeecc11531d), the MNI ICBM 152 copyright notice.






Bit-identical325,325 of 325,325 voxels equal
Axial slice 28 of 57; the crosshair where the replica and ANTs as built differ most in it; the warped T1w in the brain's 0.5–99.5 % window; the differences on one symmetric log scale over six decades below the largest, any non-zero difference tinted. The interactive viewer needs JavaScript; the metrics below are complete without it.
What separates ANTs' result at fMRIPrep's 8 threads from the replica's
- ANTs antsRegistration 2.6.5, 8 threadsas built (fMRIPrep ran 8)
- 1 thread instead of 8max |Δ| 112
- ANTs antsRegistration 2.6.5, 1 thread, as builtthe repository's build (GCC 14.3), glibc's maths
- correctly rounded maths instead of glibc'smax |Δ| 52.6
- ANTs antsRegistration 2.6.5, correctly rounded maths, 1 threadthe same binary, CORE-MATH preloaded
- replica1 and 12 threads, default and reproduce
Each box holds runs that give the same bytes, every output (the warped T1w and the three components of the displacement field); the largest |Δ| of a link is over all four. ANTs’ thread count and glibc’s maths are the only two causes, and the replica computes ANTs’ one-thread result with correctly rounded maths, on any number of threads.
Every comparison, computed in float64 over every voxel of the outputs
| Comparison (A − B), verdict and reason | Output | max |Δ| | Values that differ | RMS of Δ | max |Δ| ÷ range | 1 − NCC |
|---|---|---|---|---|---|---|
| The original against itself: the yardstick | ||||||
| ANTs antsRegistration 2.6.5, 8 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 112 | 129,599 of 325,325 (39.8 %) | 5.36 | 0.0559 | 0.000156 |
| Field x (LPS) | 0.335 | 276,789 of 325,325 (85.1 %) | 0.0816 | 0.0858 | 0.0181 | |
| Field y (LPS) | 0.494 | 276,653 of 325,325 (85 %) | 0.0937 | 0.0958 | 0.0166 | |
| Field z (LPS) | 0.319 | 276,598 of 325,325 (85 %) | 0.0558 | 0.0543 | 0.0043 | |
| ANTs antsRegistration 2.6.5 with correctly rounded maths − as built, 1 threadreferenceThe same ANTs binary, its glibc functions replaced by correctly rounded ones. platform maths | Warped T1w | 52.6 | 129,657 of 325,325 (39.9 %) | 2.6 | 0.0262 | 3.67e−5 |
| Field x (LPS) | 0.168 | 276,905 of 325,325 (85.1 %) | 0.0387 | 0.0431 | 0.00496 | |
| Field y (LPS) | 0.203 | 276,661 of 325,325 (85 %) | 0.0329 | 0.0394 | 0.00203 | |
| Field z (LPS) | 0.265 | 276,611 of 325,325 (85 %) | 0.0546 | 0.0451 | 0.00328 | |
| ANTs antsRegistration 2.6.5, 2 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 254 | 130,625 of 325,325 (40.2 %) | 11.1 | 0.127 | 0.000662 |
| Field x (LPS) | 0.234 | 276,946 of 325,325 (85.1 %) | 0.0516 | 0.06 | 0.01 | |
| Field y (LPS) | 0.403 | 276,722 of 325,325 (85.1 %) | 0.0842 | 0.0781 | 0.0147 | |
| Field z (LPS) | 0.534 | 276,731 of 325,325 (85.1 %) | 0.125 | 0.0908 | 0.0178 | |
| ANTs antsRegistration 2.6.5, 4 threads − 1 threadreferenceANTs’ result moves with its thread count: its metrics sum per-thread partial results in an order that depends on the threads. | Warped T1w | 283 | 129,997 of 325,325 (40 %) | 11.7 | 0.141 | 0.000741 |
| Field x (LPS) | 0.72 | 276,822 of 325,325 (85.1 %) | 0.145 | 0.185 | 0.0637 | |
| Field y (LPS) | 0.551 | 276,692 of 325,325 (85.1 %) | 0.113 | 0.107 | 0.0277 | |
| Field z (LPS) | 0.456 | 276,626 of 325,325 (85 %) | 0.0784 | 0.0775 | 0.0087 | |
| The replica against the original | ||||||
| replica − ANTs antsRegistration 2.6.5, 1 thread, as builtIdentical with correctly rounded mathsOnly where glibc’s exp, log, sin or cos is not correctly rounded: with CORE-MATH preloaded, the same ANTs binary gives the replica’s output bit for bit (the row against correctly rounded maths). The change then grows through the optimiser’s iterations, as ANTs’ own threads do. platform maths | Warped T1w | 52.6 | 129,657 of 325,325 (39.9 %) | 2.6 | 0.0262 | 3.67e−5 |
| Field x (LPS) | 0.168 | 276,905 of 325,325 (85.1 %) | 0.0387 | 0.0431 | 0.00496 | |
| Field y (LPS) | 0.203 | 276,661 of 325,325 (85 %) | 0.0329 | 0.0394 | 0.00203 | |
| Field z (LPS) | 0.265 | 276,611 of 325,325 (85 %) | 0.0546 | 0.0451 | 0.00328 | |
| replica − ANTs antsRegistration 2.6.5, 1 thread, correctly rounded mathsBit-identicalWith correctly rounded maths, ANTs computes exactly what the replica computes: the warped image and every value of the displacement field. | every output | identical: 0 of 325,325 values differ | ||||
| The replica against itself: thread counts, fix and reproduce | ||||||
| replica, 12 threads − 1 threadBit-identicalThe replica gives ANTs’ single-thread result on any number of threads. | every output | identical: 0 of 325,325 values differ | ||||
| replica, fix − reproduceBit-identicalNo correction of the register applies to this command: both modes give the same bytes. | every output | identical: 0 of 325,325 values differ | ||||
| replica, fix, 12 threads − 1 threadBit-identicalThe same bytes on any number of threads. | every output | identical: 0 of 325,325 values differ | ||||
Δ is A − B. Every number is computed in float64 over all 325,325 voxels of the 65 × 77 × 65 outputs (the field: each component), not only the brain shown; the range is B’s maximum minus its minimum. The rows the original against itself are ANTs’ own variability on this input. The field’s components are in mm, in ITK’s physical space (LPS).
The linear stages’ result: the affine of the composite transform (Rigid then Affine, collapsed; AffineTransform_float_3_3)
| Parameter | ANTs, 1 thread, as built | ANTs, 1 thread, CORE-MATH | replica | replica − ANTs as built | replica − ANTs with CORE-MATH | ANTs 8 threads − 1 thread |
|---|---|---|---|---|---|---|
| matrix 1,1 | 0.9440594 | 0.9426623 | 0.9426623 | −0.0014 | 0 | 0.0025 |
| matrix 1,2 | 9.248753e−4 | −2.785155e−4 | −2.785155e−4 | −0.0012 | 0 | 0.000987 |
| matrix 1,3 | 0.0035838322 | 0.0042304373 | 0.0042304373 | 0.000647 | 0 | −0.000942 |
| matrix 2,1 | 0.007971352 | 0.008520667 | 0.008520667 | 0.000549 | 0 | −0.00131 |
| matrix 2,2 | 0.8210773 | 0.82226926 | 0.82226926 | 0.00119 | 0 | 0.00305 |
| matrix 2,3 | 0.2969963 | 0.29518113 | 0.29518113 | −0.00182 | 0 | 0.000486 |
| matrix 3,1 | −0.013978851 | −0.014700439 | −0.014700439 | −0.000722 | 0 | 0.00377 |
| matrix 3,2 | −0.32897782 | −0.32976612 | −0.32976612 | −0.000788 | 0 | −0.00173 |
| matrix 3,3 | 0.75971 | 0.757824 | 0.757824 | −0.00189 | 0 | −0.00271 |
| translation x | 1.9234301 | 1.9703183 | 1.9703183 | 0.0469 | 0 | 0.135 |
| translation y | 11.570224 | 11.559067 | 11.559067 | −0.0112 | 0 | 0.102 |
| translation z | −17.417015 | −17.481945 | −17.481945 | −0.0649 | 0 | −0.0966 |
The parameters as the files store them (float32: antsRegistration ran with --float 1), printed with the digits that identify each value; the matrix row by row in ITK’s physical space (LPS), the translation in mm. The centre of rotation is the same in every run (0.250445, 21.9911, 9.73538 mm).
The displacement fields: how far a voxel’s displacement moves (the length of the difference vector, mm), over the template’s brain
| Comparison | largest (mm) | 99th percentile (mm) | RMS (mm) | voxels that differ, of 69,880 |
|---|---|---|---|---|
| replica − ANTs as built, 1 thread | 0.271 | 0.25 | 0.123 | 69,880 |
| replica − ANTs with CORE-MATH, 1 thread | 0 | 0 | 0 | 0 |
| ANTs 2 threads − 1 thread | 0.64 | 0.558 | 0.255 | 69,880 |
| ANTs 4 threads − 1 thread | 0.922 | 0.788 | 0.328 | 69,880 |
| ANTs 8 threads − 1 thread | 0.501 | 0.457 | 0.235 | 69,880 |
| ANTs with CORE-MATH − as built, 1 thread | 0.271 | 0.25 | 0.123 | 69,880 |
ANTs’ own runs at 2, 4 and 8 threads are the yardstick: the replica’s difference from ANTs as built is the size of a change of ANTs’ thread count, and nothing with correctly rounded maths.
Provenance of this example
- Data: OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01, PDDL-1.0; MNI ICBM 152 nonlinear asymmetric 2009c (Fonov et al., 2009, 2011), from TemplateFlow, the MNI ICBM 152 notice (static/hero/LICENSE-MNI.txt)
- Inputs: fixed (sha256 b06d202af1954dbf…); moving (sha256 d358573fa558039d…)
- Runs: figures cache entry
ants-registration/fmriprep-ds000005-template-tvf-small, computed 2026-10-22T23:50:15+00:00 with larmorxa9401a69f759; ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0; every command, its environment, threads and exit status are in itsRUN_INFO.json - Stage under test: TimeVaryingVelocityField[0.5,4,3,0,0,0]
Every number above is in ants-registration.json, exported by python -m larmorx_validation evidence ants-registration.
Against the original’s own variability
ANTs’ result moves with its thread count: its metrics add up partial sums per thread, in an order that depends on the threads. The record measures this variability, ANTs at 2, 4 and 8 threads against its own one-thread result, and gives for each case the replica’s distance from ANTs at one thread.
antsRegistration 2.6.5, the largest change in a transform parameter in each of the 36 cases (log scale; 0 at the left edge)- the original against itself: 2, 4 and 8 threads against 1
- the replica against the original: ANTs at 1 thread, as built
smoke/rigidthe original against itself: 2, 4 and 8 threads against 1 6.96e−13the replica against the original: ANTs at 1 thread, as built 0smoke/affine-floatthe original against itself: 2, 4 and 8 threads against 1 0.00488the replica against the original: ANTs at 1 thread, as built 0metrics/mattes-mi-aliasthe original against itself: 2, 4 and 8 threads against 1 9.08e−13the replica against the original: ANTs at 1 thread, as built 0metrics/mi2the original against itself: 2, 4 and 8 threads against 1 3.01e−14the replica against the original: ANTs at 1 thread, as built 3.47e−18metrics/mi2-densethe original against itself: 2, 4 and 8 threads against 1 9.71e−17the replica against the original: ANTs at 1 thread, as built 0metrics/meansquaresthe original against itself: 2, 4 and 8 threads against 1 8.67e−14the replica against the original: ANTs at 1 thread, as built 0metrics/gcthe original against itself: 2, 4 and 8 threads against 1 1.61e−12the replica against the original: ANTs at 1 thread, as built 0metrics/gc-densethe original against itself: 2, 4 and 8 threads against 1 1.45e−13the replica against the original: ANTs at 1 thread, as built 0metrics/cc-sampledthe original against itself: 2, 4 and 8 threads against 1 2.47e−13the replica against the original: ANTs at 1 thread, as built 0metrics/cc-densethe original against itself: 2, 4 and 8 threads against 1 5.47e−14the replica against the original: ANTs at 1 thread, as built 0metrics/mattes-gradient-filterthe original against itself: 2, 4 and 8 threads against 1 4.37e−16the replica against the original: ANTs at 1 thread, as built 0metrics/cc-gradient-filterthe original against itself: 2, 4 and 8 threads against 1 9.42e−14the replica against the original: ANTs at 1 thread, as built 0point-sets/icpthe original against itself: 2, 4 and 8 threads against 1 0the replica against the original: ANTs at 1 thread, as built 0point-sets/psethe original against itself: 2, 4 and 8 threads against 1 1.11e−16the replica against the original: ANTs at 1 thread, as built 0point-sets/jhctthe original against itself: 2, 4 and 8 threads against 1 1.11e−16the replica against the original: ANTs at 1 thread, as built 0point-sets/igdmthe original against itself: 2, 4 and 8 threads against 1 2.12e−22the replica against the original: ANTs at 1 thread, as built 0syn/ccthe original against itself: 2, 4 and 8 threads against 1 0the replica against the original: ANTs at 1 thread, as built 0syn/meansquares-totalthe original against itself: 2, 4 and 8 threads against 1 0the replica against the original: ANTs at 1 thread, as built 0syn/mattesthe original against itself: 2, 4 and 8 threads against 1 3.55e−14the replica against the original: ANTs at 1 thread, as built 2.13e−14syn-point-sets/icpthe original against itself: 2, 4 and 8 threads against 1 0.0543the replica against the original: ANTs at 1 thread, as built 0bspline-syn/ccthe original against itself: 2, 4 and 8 threads against 1 2.33e−6the replica against the original: ANTs at 1 thread, as built 0bspline-syn/knot-spacingthe original against itself: 2, 4 and 8 threads against 1 6.65e−7the replica against the original: ANTs at 1 thread, as built 0displacement-field/gaussian-ccthe original against itself: 2, 4 and 8 threads against 1 0the replica against the original: ANTs at 1 thread, as built 0displacement-field/bspline-meansquaresthe original against itself: 2, 4 and 8 threads against 1 8.87e−7the replica against the original: ANTs at 1 thread, as built 0bspline/meansquaresthe original against itself: 2, 4 and 8 threads against 1 8.88e−16the replica against the original: ANTs at 1 thread, as built 0exponential/meansquaresthe original against itself: 2, 4 and 8 threads against 1 0the replica against the original: ANTs at 1 thread, as built 0bspline-exponential/meansquaresthe original against itself: 2, 4 and 8 threads against 1 3.21e−8the replica against the original: ANTs at 1 thread, as built 0time-varying/meansquaresthe original against itself: 2, 4 and 8 threads against 1 0the replica against the original: ANTs at 1 thread, as built 0time-varying-bspline/meansquaresthe original against itself: 2, 4 and 8 threads against 1 1.91e−6the replica against the original: ANTs at 1 thread, as built 0real/ds000005-sub-01/boldref-affinethe original against itself: 2, 4 and 8 threads against 1 9.95e−10the replica against the original: ANTs at 1 thread, as built 0real/ds005040-sub-002/coregisterthe original against itself: 2, 4 and 8 threads against 1 1.68e−12the replica against the original: ANTs at 1 thread, as built 0real-syn-small/ds000005-sub-01/brain-extractionthe original against itself: 2, 4 and 8 threads against 1 1.7the replica against the original: ANTs at 1 thread, as built 1.63real-syn-small/ds000005-sub-01/templatethe original against itself: 2, 4 and 8 threads against 1 1.45the replica against the original: ANTs at 1 thread, as built 0.945real-bsplinesyn-small/ds000005-sub-01/templatethe original against itself: 2, 4 and 8 threads against 1 1.17the replica against the original: ANTs at 1 thread, as built 0.51real-exponential-small/ds000005-sub-01/templatethe original against itself: 2, 4 and 8 threads against 1 0.39the replica against the original: ANTs at 1 thread, as built 0.113real-time-varying-small/ds000005-sub-01/templatethe original against itself: 2, 4 and 8 threads against 1 0.603the replica against the original: ANTs at 1 thread, as built 0.265
From the record: its thread-band sentence and each case’s reported distance. Where the replica differs from ANTs as built, ANTs with correctly rounded maths gives the replica’s result exactly.
The corrections, and what they change
Where ANTs has a deterministic defect that changes a result, the replica corrects it by default, fix, and warns. upstream_defects="reproduce" (on the command line --upstream-defects reproduce from the next release, LARMORX_UPSTREAM_DEFECTS=reproduce in 0.1.0rc1) gives ANTs’ behaviour, and every comparison above runs with it. The record’s list of what fix changed names, case by case, the corrections a command meets and what they change; none applies to sMRIPrep’s commands above. The corrections change the point-set metrics, the B-spline field meshes and what can follow a B-spline transform; each was measured on phantoms whose answer is known, or against the closest runs ANTs can do (findings).
IGDM (ANTS-67, ANTS-68, ANTS-70): a rigid registration of the point-set phantoms, whose moving blocks are shifted by 3.77 mm; the translation error each mode leaves
| Case | ANTs (reproduce): error (mm) | corrected: error (mm) | ANTs: iterations | corrected: iterations |
|---|---|---|---|---|
| given sigmas, one neighbour, central differences | 3.77 | 3.3 | 40 | 40 |
| one neighbour, gradientSigma 1 mm | 3.77 | 3.28 | 40 | 40 |
| kNeighborhood 5 | 3.77 | 0.921 | 40 | 40 |
| kNeighborhood 5, gradientSigma 1.5 mm | 3.77 | 0.623 | 40 | 40 |
| estimated sigmas (the floor applies) | 3.77 | 3.33 | 40 | 40 |
| an image with itself, half a voxel apart | 1 | 0.000703 | 9 | 10 |
From the ants-registration-corrections record (the replica in fix, its default, against reproduce, which the ants-registration record compares with ANTs bit for bit; for the B-spline chains, against ANTs 2.6.5 itself).
JHCT (ITK-24, ITK-29): the same registration with the labelled phantoms
| Case | ANTs (reproduce): error (mm) | corrected: error (mm) | largest parameter change |
|---|---|---|---|
| isotropic, sigma 2: the total alone (ITK-24) | 0.0345 | 0.0345 | 1.1e−5 |
| isotropic, sigma 1 (ITK-24, ITK-23) | 5.59e−6 | 4.36e−5 | 4.32e−5 |
| anisotropic, sigma 1 (ITK-29, ITK-24, ITK-26) | 0.0273 | 0.0101 | 0.0178 |
| anisotropic, sigma 2, alpha 1.5 | 0.104 | 0.0583 | 0.0357 |
From the ants-registration-corrections record (the replica in fix, its default, against reproduce, which the ants-registration record compares with ANTs bit for bit; for the B-spline chains, against ANTs 2.6.5 itself).
B-spline field meshes at the base level (ANTS-78; ITK-42): ANTs’ stage is the corrected one with twice the mesh, byte for byte; how far the corrected field moves from ANTs’ (mm, in the mask) and the fit each leaves
| Case | field moved: largest (mm) | mean (mm) | ANTs' fit (MSE) | corrected fit (MSE) |
|---|---|---|---|---|
| BSplineDisplacementField 3x3x3, two levels | 4.16 | 1.43 | 3.01e4 | 3.01e4 |
| 5x5x5, total mesh 4x4x4 | 5.48 | 2.12 | 3.11e4 | 3.56e4 |
| a knot spacing of 12 mm | 3.04 | 1.09 | 2.92e4 | 2.97e4 |
| BSplineExponential 4x4x4 | 0.12 | 0.0439 | 3.75e4 | 3.8e4 |
From the ants-registration-corrections record (the replica in fix, its default, against reproduce, which the ants-registration record compares with ANTs bit for bit; for the B-spline chains, against ANTs 2.6.5 itself).
ANTs cannot run a stage after a B-spline transform (ITK-35), so the last table sets the corrected stage against the two closest runs ANTs can do: the same stage on the moving image resampled through the B-spline transform, and ANTs given the B-spline transform sampled as a displacement field (at half its control-point spacing) as its initial transform. The field is close for two B-spline stages (a quarter of a millimetre: the field’s sampling). For a linear stage it is not, because ITK takes a field’s Jacobian from its nearest voxel and the linear stage stops early; the replica given the same field stops with it, identically. That reference also varies from run to run: a linear stage after a displacement field estimates its parameter scales and learning rate from 1000 voxels that ITK draws with a generator seeded from the clock, whatever the seed (ANTS-72; with time and clock fixed by a preloaded library, every run is identical). The record therefore runs it five times and gives the median, and the last column how far ANTs’ own runs are apart.
Stages after a B-spline transform (ITK-35), which ANTs cannot run: the corrected stage against ANTs’ closest runs, how far their results map the fixed points apart (largest and mean, mm, in the mask); with the field, the median of five ANTs runs, and how far ANTs’ own runs are apart (ANTS-72)
| Case | vs ANTs on the resampled image: largest | mean | vs ANTs with the field (half spacing): largest | mean | ANTs' own runs apart: largest |
|---|---|---|---|---|---|
| BSpline, then Affine | 0.366 | 0.185 | 4.66 | 2.34 | 3.2 |
| BSpline, then Rigid | 0.292 | 0.143 | 3.12 | 1.86 | 7.86 |
| -r a B-spline transform, then Rigid | 2.63 | 1.46 | 4.58 | 2.44 | 0.000757 |
| -r a B-spline transform, then Affine | 7.54 | 3.96 | 9.27 | 4.24 | 2.99 |
| two BSpline stages | 0.838 | 0.44 | 0.238 | 0.129 | 0 |
From the ants-registration-corrections record (the replica in fix, its default, against reproduce, which the ants-registration record compares with ANTs bit for bit; for the B-spline chains, against ANTs 2.6.5 itself).
A reference that varies: for the linear stages after a B-spline transform, read the “with the field” columns as the median of a varying reference, next to its own spread: ANTs’ runs with the field as initial transform are several millimetres apart from each other (the last column), as far as they are from the replica. The replica’s results and the reference on the resampled image are identical from run to run.
On the data panel
The data panel is the fixed set of fourteen OpenNeuro subjects on which LARMORX’s programs are validated: 1.5, 3 and 7 T; Siemens, GE and Philips; single-band, multiband and multi-echo BOLD; a 3.5-year-old to an 82-year-old; a tumour, a stroke with its lesion mask, strong motion. Its record replays every call fMRIPrep 25.2.6 made of the program in its fourteen runs on these subjects, the node’s own command.txt with its inputs from the run’s working directory (calls repeated word for word are compared once), and runs the program on panel images fMRIPrep never gives it. ANTs runs at one thread, as built and with correctly rounded maths, and at 2 and 3 threads for its own variability where it has any; the replica in reproduce, and in fix for the corrections. The panel’s cases are in the case explorer below (the categories data panel: fMRIPrep’s calls and data panel: fMRIPrep’s calls, every third voxel).
71 of 71 compared cases identical to ANTs with correctly rounded maths, 16 of them as built: every antsRegistration node of the fourteen runs with fMRIPrep’s seed (the BOLD references’ registrations to the template, the field maps’ coregistrations, sMRIPrep’s brain extraction at full size) and the whole template registrations on every third voxel. At full size, about 2.4 hours of ANTs at one thread each, the fourteen template registrations have their own record: 14 of 14 identical to ANTs with correctly rounded maths. The template registration’s inputs, which niworkflows writes for itself and nipype does not keep, are made again exactly from each node’s recorded inputs. In fix the replica corrects ITK-31 in sMRIPrep’s brain extractions, whose SyN stage has a fixed mask only: the panel’s record names it in 28 cases, and their values change (register).
fMRIPrep’s antsRegistration nodes in the fourteen fMRIPrep runs, and what the panel record compares of them
| Node (normalised path) | Nodes | Runs | Compared | Same call as a compared one | Statuses | Not compared: why |
|---|---|---|---|---|---|---|
| anat_fit_wf/brain_extraction_wf/norm | 14 | 14 | 14 | 0 | 14 pass | – |
| anat_fit_wf/register_template_wf/registration | 14 | 14 | 0 | 0 | – | full size not run in this record (hours of ANTs at one thread each, as built and with CORE-MATH: the next part, on a larger machine); compared on every 3rd voxel instead (28 small cases) |
| bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm | 22 | 12 | 22 | 0 | 22 pass | – |
| bold_wf/bold_fit_wf/fmapreg_wf/coregister | 7 | 4 | 7 | 0 | 7 pass | – |
From the panel record (docs/validation/ants-registration-panel.md): every node of the finished fMRIPrep 25.2.6 runs that ran the program, grouped by its path with the subject, run, space and item removed; a call repeated word for word is compared once.
Images on a small grid (a BOLD reference, a template at 2 mm) are shown as whole volumes; the panel’s large anatomical images (up to 17.5 million voxels) as three full-resolution slices through the voxel where the replica and ANTs as built differ most (the centre of the field of view where they agree), the whole field of view shown, defaced heads included (the data panel’s defacing check).
Worked example
Data panel: fMRIPrep's BOLD reference to MNI at 7 T
What this shows: In reproduce, on 1 and 8 threads, the replica gives ANTs’ bytes with correctly rounded maths; ANTs as built differs only where glibc’s maths is not correctly rounded (largest difference 2.1e-08). No correction by default changes this call.
- Replica against ANTs antsRegistration 2.6.5Identical with correctly rounded maths
fMRIPrep 25.2.6’s own call (the BOLD reference’s registration to MNI152NLin2009cAsym’s fMRIPrep BOLD reference at 2 mm: Affine, Mattes with random sampling from fMRIPrep’s seed) on ds005454 sub-16’s 7 T multiband BOLD reference (1.4 mm): each run’s transform places the BOLD reference on the template’s grid (ANTs’ antsApplyTransforms, linear). A key dataset, shown as a whole volume; every run’s parameters are in the table below it.
- Data
- OpenNeuro ds005454, sub-16 (CC-BY-4.0); the whole field of view shown, only margins where every array is 0 left out
- Node
bold[task_rest]/bold_fit/enhance_and_skullstrip_bold/normof the fMRIPrep run ds005454-sub-16-noreconall- Command
antsRegistration --collapse-output-transforms 1 --dimensionality 3 --initial-moving-transform [ initialization.mat, 0 ] --initialize-transforms-per-stage 0 --interpolation Linear --output transform --transform Affine[ 0.1 ] --metric Mattes[ tpl-MNI152NLin2009cAsym_res-02_desc-fMRIPrep_boldref.nii.gz, sub-16_task-rest_desc-hmc_boldref.nii.gz, 1, 64, Random, 0.2 ] --convergence [ 200, 1e-09, 10 ] --smoothing-sigmas 2.0mm --shrink-factors 2 --use-histogram-matching 1 --winsorize-image-intensities [ 0.05, 0.98 ] --write-composite-transform 0- Programs
- ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0; the replica at commit
042d70d5a17c - Grid
- 97 × 115 × 97 voxels of 2 × 2 × 2 mm; shown 97 × 115 × 92, every slice
Data shown
- fMRIPrep boot camp 7T training dataset (OpenNeuro ds005454, version 1.0.0, doi:10.18112/openneuro.ds005454.v1.0.0), Creative Commons Attribution 4.0 (CC BY 4.0). Attribution: Sleight EC, Esteban O, Grouiller F. fMRIPrep boot camp -- 7T training dataset. OpenNeuro ds005454, version 1.0.0. doi:10.18112/openneuro.ds005454.v1.0.0 The images shown are adapted from it: reoriented to RAS, cropped to the region shown, masked outside it, and shown as the outputs of the programs compared.
- MNI152NLin2009cAsym template (TemplateFlow, version tpl-MNI152NLin2009cAsym git 15d7c02160f79f5218d2545b4febebeecc11531d), the MNI ICBM 152 copyright notice.





Bit-identical1,082,035 of 1,082,035 voxels equal
Bit-identical1,082,035 of 1,082,035 voxels equal
Axial slice 46 of 92; the crosshair where the replica differs most from ANTs as built. The interactive viewer needs JavaScript; the metrics below are complete without it.
Every comparison, computed in float64 over every voxel of the outputs
| Comparison (A − B), verdict and reason | max |Δ| | Values that differ | RMS of Δ | max |Δ| ÷ range | 1 − NCC |
|---|---|---|---|---|---|
| The original against itself: the yardstick | |||||
| ANTs antsRegistration 2.6.5, 8 threads − 1 threadreferenceANTs’ own result moves with its thread count (placed): the yardstick for any difference above. | 1.87e−6 | 896,328 of 1,082,035 (82.8 %) | 6.01e−8 | 1.81e−9 | 6.63e−20 |
| The replica against the original | |||||
| replica − ANTs antsRegistration 2.6.5, 1 thread, as builtIdentical with correctly rounded mathsOnly where glibc’s maths is not correctly rounded (placed): with CORE-MATH preloaded, ANTs gives the replica’s bytes. | 2.1e−8 | 896,318 of 1,082,035 (82.8 %) | 7.5e−10 | 2.04e−11 | 1.03e−23 |
| replica − ANTs antsRegistration 2.6.5, correctly rounded mathsBit-identicalThe replica gives ANTs’ bytes with correctly rounded maths. | identical: 0 of 1,082,035 values differ | ||||
| The replica against itself: thread counts, fix and reproduce | |||||
| replica, fix − reproduceBit-identicalNo correction of the defect register changes this call. | identical: 0 of 1,082,035 values differ | ||||
| replica, 1 thread − 8 threads (reproduce)Bit-identicalThe replica’s result does not depend on its thread count. | identical: 0 of 1,082,035 values differ | ||||
Δ is A − B over all 1,082,035 values of each output (the moving image placed on the fixed grid through each run’s transform).
Provenance of this example
- Data: OpenNeuro ds005454, sub-16, CC-BY-4.0
- Inputs: input 1 (sha256 c6524103102987da…)
- Runs: figures cache entry
ants-registration/panel-fmriprep-boldref-7t, computed 2026-10-27T01:40:47+00:00 with larmorx042d70d5a17c; ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0; every command, its environment, threads and exit status are in itsRUN_INFO.json
Every number above is in ants-registration.json, exported by python -m larmorx_validation evidence ants-registration.
Data panel: fMRIPrep’s BOLD reference to MNI at 7 T: every run’s transform
| Run | Class | p0 | p1 | p2 | p3 | p4 | p5 | p6 | p7 | p8 | p9 | p10 | p11 | Centre (fixed parameters) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ANTs antsRegistration 2.6.51 thread · as built (GCC 14.3), glibc’s maths | AffineTransform_double_3_3 | 0.8695858586795301 | 0.006695985014170093 | 4.951559045512054e−4 | −0.03494006259232191 | 0.8305532382266546 | 0.3336756611818985 | 0.008088924657355695 | −0.32833462642056066 | 0.7578730003602518 | 2.039522862293599 | −9.380221859996066 | −10.241086502425679 | 0.218475699 20.3963718 12.6540527 |
| ANTs antsRegistration 2.6.5, correctly rounded maths1 thread, CORE-MATH preloaded | AffineTransform_double_3_3 | 0.8695858586795995 | 0.006695985014134464 | 4.951559045407467e−4 | −0.03494006259244027 | 0.8305532382266101 | 0.3336756611819881 | 0.008088924657254387 | −0.32833462642065037 | 0.7578730003606966 | 2.0395228622856614 | −9.380221860003642 | −10.2410865024278 | 0.218475699 20.3963718 12.6540527 |
| ANTs antsRegistration 2.6.58 threads · as built | AffineTransform_double_3_3 | 0.8695858586787719 | 0.006695985016565874 | 4.951559065171345e−4 | −0.0349400625849211 | 0.8305532382300399 | 0.3336756611759714 | 0.008088924661362074 | −0.32833462641490246 | 0.7578730003268729 | 2.039522862663477 | −9.380221859486497 | −10.24108650285752 | 0.218475699 20.3963718 12.6540527 |
| replicalarmorx, Apache-2.0 · reproduce, 8 threads | AffineTransform_double_3_3 | 0.8695858586795995 | 0.006695985014134464 | 4.951559045407467e−4 | −0.03494006259244027 | 0.8305532382266101 | 0.3336756611819881 | 0.008088924657254387 | −0.32833462642065037 | 0.7578730003606966 | 2.0395228622856614 | −9.380221860003642 | −10.2410865024278 | 0.218475699 20.3963718 12.6540527 |
| replicalarmorx, Apache-2.0 · reproduce, 1 thread | AffineTransform_double_3_3 | 0.8695858586795995 | 0.006695985014134464 | 4.951559045407467e−4 | −0.03494006259244027 | 0.8305532382266101 | 0.3336756611819881 | 0.008088924657254387 | −0.32833462642065037 | 0.7578730003606966 | 2.0395228622856614 | −9.380221860003642 | −10.2410865024278 | 0.218475699 20.3963718 12.6540527 |
| replicalarmorx, Apache-2.0 · fix (the default), 8 threads | AffineTransform_double_3_3 | 0.8695858586795995 | 0.006695985014134464 | 4.951559045407467e−4 | −0.03494006259244027 | 0.8305532382266101 | 0.3336756611819881 | 0.008088924657254387 | −0.32833462642065037 | 0.7578730003606966 | 2.0395228622856614 | −9.380221860003642 | −10.2410865024278 | 0.218475699 20.3963718 12.6540527 |
Every parameter as stored (the shortest float64 that reads back exactly).
Worked example
Data panel: fMRIPrep's field map to a Philips multiband BOLD reference
What this shows: In reproduce, on 1 and 8 threads, the replica gives ANTs’ bytes, as built and with correctly rounded maths. No correction by default changes this call.
- Replica against ANTs antsRegistration 2.6.5Bit-identical
fMRIPrep 25.2.6’s own call (sdcflows’ coregistration of the field map’s magnitude and the BOLD reference: two Rigid stages, Mattes with random sampling, masks) on ds002603 sub-01’s multiband-4 run (Philips, 588 volumes): each run’s transform places the BOLD reference on the magnitude’s grid.
- Data
- OpenNeuro ds002603, sub-01 (CC0-1.0); the whole field of view shown, only margins where every array is 0 left out
- Node
bold[task_wm_acq_Mb4Mm3Tr550]/bold_fit/fmapreg/coregisterof the fMRIPrep run ds002603-sub-01-noreconall- Command
antsRegistration --collapse-output-transforms 1 --dimensionality 3 --initial-moving-transform [ sub-01_acq-Mm2_fmapid-auto00000_desc-magnitude_fieldmap.nii.gz, sub-01_task-wm_acq-Mb4Mm3Tr550_desc-hmc_boldref_corrected.nii.gz, 1 ] --initialize-transforms-per-stage 0 --interpolation LanczosWindowedSinc --output transform --transform Rigid[ 0.1 ] --metric Mattes[ sub-01_acq-Mm2_fmapid-auto00000_desc-magnitude_fieldmap.nii.gz, sub-01_task-wm_acq-Mb4Mm3Tr550_desc-hmc_boldref_corrected.nii.gz, 1, 32, Random, 0.25 ] --convergence [ 50, 1e-07, 5 ] --smoothing-sigmas 8.0mm --shrink-factors 2 --use-histogram-matching 1 --masks [ sub-01_acq-Mm2_fmapid-auto00000_desc-brain_mask_dil.nii.gz, sub-01_task-wm_acq-Mb4Mm3Tr550_desc-hmc_boldref_corrected_brain_mask_maths_dil.nii.gz ] --transform Rigid[ 0.1 ] --metric Mattes[ sub-01_acq-Mm2_fmapid-auto00000_desc-magnitude_fieldmap.nii.gz, sub-01_task-wm_acq-Mb4Mm3Tr550_desc-hmc_boldref_corrected.nii.gz, 1, 32, Random, 0.5 ] --convergence [ 20, 1e-08, 3 ] --smoothing-sigmas 2.0mm --shrink-factors 1 --use-histogram-matching 1 --masks [ sub-01_acq-Mm2_fmapid-auto00000_desc-brain_mask_dil.nii.gz, sub-01_task-wm_acq-Mb4Mm3Tr550_desc-hmc_boldref_corrected_brain_mask_maths_dil.nii.gz ] --winsorize-image-intensities [ 0.001, 0.999 ] --write-composite-transform 0- Programs
- ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0; the replica at commit
042d70d5a17c - Grid
- 128 × 128 × 128 voxels of 2 × 2 × 2 mm; shown 126 × 128 × 127, three full-resolution slices
Data shown
- MB-EPI-comparison (OpenNeuro ds002603, version 1.0.0, doi:10.18112/openneuro.ds002603.v1.0.0), CC0 1.0 public-domain dedication.




Bit-identical2,097,152 of 2,097,152 voxels equal
Bit-identical2,097,152 of 2,097,152 voxels equal
Bit-identical2,097,152 of 2,097,152 voxels equal
Axial slice 63 of 127; the crosshair at the centre of the field of view; the three slices through it are stored at full resolution. The interactive viewer needs JavaScript; the metrics below are complete without it.
Stored for this example: three full-resolution slices, one per plane, through display voxel (63, 64, 63): where the main comparison differs most, or the field's centre where nothing differs; every metric is computed over every voxel of the files.
Every comparison, computed in float64 over every voxel of the outputs
| Comparison (A − B), verdict and reason | max |Δ| | Values that differ | RMS of Δ | max |Δ| ÷ range | 1 − NCC |
|---|---|---|---|---|---|
| The original against itself: the yardstick | |||||
| ANTs antsRegistration 2.6.5, 8 threads − 1 threadreferenceANTs’ own result moves with its thread count (placed): the yardstick for any difference above. | 9.22e−8 | 853,957 of 2,097,152 (40.7 %) | 2.74e−9 | 1.21e−12 | 2.38e−26 |
| The replica against the original | |||||
| replica − ANTs antsRegistration 2.6.5, 1 thread, as builtBit-identicalThe replica gives the unmodified ANTs program’s bytes. | identical: 0 of 2,097,152 values differ | ||||
| replica − ANTs antsRegistration 2.6.5, correctly rounded mathsBit-identicalThe replica gives ANTs’ bytes with correctly rounded maths. | identical: 0 of 2,097,152 values differ | ||||
| The replica against itself: thread counts, fix and reproduce | |||||
| replica, fix − reproduceBit-identicalNo correction of the defect register changes this call. | identical: 0 of 2,097,152 values differ | ||||
| replica, 1 thread − 8 threads (reproduce)Bit-identicalThe replica’s result does not depend on its thread count. | identical: 0 of 2,097,152 values differ | ||||
Δ is A − B over all 2,097,152 values of each output (the moving image placed on the fixed grid through each run’s transform).
Provenance of this example
- Data: OpenNeuro ds002603, sub-01, CC0-1.0
- Inputs: input 1 (sha256 5211a3e1d6558e9f…)
- Runs: figures cache entry
ants-registration/panel-fmriprep-coregister-philips, computed 2026-10-27T01:42:06+00:00 with larmorx042d70d5a17c; ANTs v2.6.5 (fdce4d2f84) on ITK v5.4.5 (f51594ad88), x86_64-conda-linux-gnu-g++ (conda-forge gcc 14.3.0-20) 14.3.0; every command, its environment, threads and exit status are in itsRUN_INFO.json
Every number above is in ants-registration.json, exported by python -m larmorx_validation evidence ants-registration.
Data panel: fMRIPrep’s field map to a Philips multiband BOLD reference: every run’s transform
| Run | Class | p0 | p1 | p2 | p3 | p4 | p5 | p6 | p7 | p8 | p9 | p10 | p11 | Centre (fixed parameters) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ANTs antsRegistration 2.6.51 thread · as built (GCC 14.3), glibc’s maths | AffineTransform_double_3_3 | 0.9998618205985751 | −0.012886477506775495 | −0.01050135261622249 | 0.012970196402157963 | 0.9998843309028401 | 0.007943476585784513 | 0.010397774501898108 | −0.008078583566865378 | 0.9999133076286968 | −2.0935394254810404 | −3.1515078852597957 | 3.7083510016714083 | 0.584962374 -2.25883554 -10.259005 |
| ANTs antsRegistration 2.6.5, correctly rounded maths1 thread, CORE-MATH preloaded | AffineTransform_double_3_3 | 0.9998618205985751 | −0.012886477506775495 | −0.01050135261622249 | 0.012970196402157963 | 0.9998843309028401 | 0.007943476585784513 | 0.010397774501898108 | −0.008078583566865378 | 0.9999133076286968 | −2.0935394254810404 | −3.1515078852597957 | 3.7083510016714083 | 0.584962374 -2.25883554 -10.259005 |
| ANTs antsRegistration 2.6.58 threads · as built | AffineTransform_double_3_3 | 0.9998618205985739 | −0.012886477506831383 | −0.010501352616264732 | 0.012970196402214617 | 0.999884330902839 | 0.0079434765858254 | 0.010397774501939364 | −0.008078583566907395 | 0.999913307628696 | −2.0935394254807034 | −3.1515078852597815 | 3.7083510016710157 | 0.584962374 -2.25883554 -10.259005 |
| replicalarmorx, Apache-2.0 · reproduce, 8 threads | AffineTransform_double_3_3 | 0.9998618205985751 | −0.012886477506775495 | −0.01050135261622249 | 0.012970196402157963 | 0.9998843309028401 | 0.007943476585784513 | 0.010397774501898108 | −0.008078583566865378 | 0.9999133076286968 | −2.0935394254810404 | −3.1515078852597957 | 3.7083510016714083 | 0.584962374 -2.25883554 -10.259005 |
| replicalarmorx, Apache-2.0 · reproduce, 1 thread | AffineTransform_double_3_3 | 0.9998618205985751 | −0.012886477506775495 | −0.01050135261622249 | 0.012970196402157963 | 0.9998843309028401 | 0.007943476585784513 | 0.010397774501898108 | −0.008078583566865378 | 0.9999133076286968 | −2.0935394254810404 | −3.1515078852597957 | 3.7083510016714083 | 0.584962374 -2.25883554 -10.259005 |
| replicalarmorx, Apache-2.0 · fix (the default), 8 threads | AffineTransform_double_3_3 | 0.9998618205985751 | −0.012886477506775495 | −0.01050135261622249 | 0.012970196402157963 | 0.9998843309028401 | 0.007943476585784513 | 0.010397774501898108 | −0.008078583566865378 | 0.9999133076286968 | −2.0935394254810404 | −3.1515078852597957 | 3.7083510016714083 | 0.584962374 -2.25883554 -10.259005 |
Every parameter as stored (the shortest float64 that reads back exactly).
Every case
Each case of the two records, the standard comparison and the data panel, with what it tests, the command line, its verdict and the reason. Search, filter by category or verdict (or select a segment of the chart above), sort by a column; #case=<name> links to one case.
584 cases
| Case | Category | What it tests | Verdict | Checks |
|---|---|---|---|---|
smoke/rigid | first checks | Rigid, Mattes, regular sampling, 2 levels, warped imagescommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 9/9 |
smoke/affine-float | first checks | Affine, MeanSquares, random sampling, --float 1command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] --float 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,0,Random,0.3]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 9/9 |
smoke/composite | first checks | -a 1: the composite and its inverse (HDF5)command lineantsRegistration -d 3 -v 1 -o r -a 1 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Translation-3d | linear transforms | Translation in 3Dcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Translation[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
transforms/Translation-3d-z0 | linear transforms | Translation in 3D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o r -z 0 -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Translation[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Translation-3d-float | linear transforms | Translation in 3D, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t Translation[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Translation-2d | linear transforms | Translation in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Translation[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Translation-2d-z0 | linear transforms | Translation in 2D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 2 -v 1 -o r -z 0 -r '[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -t Translation[0.1] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,0,Regular,0.5]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Translation-2d-float | linear transforms | Translation in 2D, --float 1command lineantsRegistration -d 2 -v 1 -o r --float 1 -t Translation[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Random,0.5]' -c [10,1e-6,5] -f 1 -s 1 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Translation-4d | linear transforms | Translation in 4D (an affine transform in 4D)command lineantsRegistration -d 4 -v 1 -o r -t Translation[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Rigid-3d | linear transforms | Rigid in 3Dcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
transforms/Rigid-3d-z0 | linear transforms | Rigid in 3D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o r -z 0 -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Rigid-3d-float | linear transforms | Rigid in 3D, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t Rigid[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Rigid-2d | linear transforms | Rigid in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Rigid-2d-z0 | linear transforms | Rigid in 2D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 2 -v 1 -o r -z 0 -r '[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -t Rigid[0.1] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,0,Regular,0.5]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Rigid-2d-float | linear transforms | Rigid in 2D, --float 1command lineantsRegistration -d 2 -v 1 -o r --float 1 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Random,0.5]' -c [10,1e-6,5] -f 1 -s 1 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Rigid-4d | linear transforms | Rigid in 4D (an affine transform in 4D)command lineantsRegistration -d 4 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
transforms/Similarity-3d | linear transforms | Similarity in 3Dcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Similarity[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
transforms/Similarity-3d-z0 | linear transforms | Similarity in 3D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o r -z 0 -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Similarity[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Similarity-3d-float | linear transforms | Similarity in 3D, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t Similarity[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Similarity-2d | linear transforms | Similarity in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Similarity[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Similarity-2d-z0 | linear transforms | Similarity in 2D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 2 -v 1 -o r -z 0 -r '[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -t Similarity[0.1] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,0,Regular,0.5]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Similarity-2d-float | linear transforms | Similarity in 2D, --float 1command lineantsRegistration -d 2 -v 1 -o r --float 1 -t Similarity[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Random,0.5]' -c [10,1e-6,5] -f 1 -s 1 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Similarity-4d | linear transforms | Similarity in 4D (an affine transform in 4D)command lineantsRegistration -d 4 -v 1 -o r -t Similarity[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
transforms/Affine-3d | linear transforms | Affine in 3Dcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
transforms/Affine-3d-z0 | linear transforms | Affine in 3D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o r -z 0 -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in r1Affine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
transforms/Affine-3d-float | linear transforms | Affine in 3D, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Affine-2d | linear transforms | Affine in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Affine-2d-z0 | linear transforms | Affine in 2D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 2 -v 1 -o r -z 0 -r '[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,0,Regular,0.5]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/Affine-2d-float | linear transforms | Affine in 2D, --float 1command lineantsRegistration -d 2 -v 1 -o r --float 1 -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Random,0.5]' -c [10,1e-6,5] -f 1 -s 1 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/Affine-4d | linear transforms | Affine in 4D (an affine transform in 4D)command lineantsRegistration -d 4 -v 1 -o r -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
transforms/CompositeAffine-3d | linear transforms | CompositeAffine in 3Dcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t CompositeAffine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, r0GenericAffine.mat, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
transforms/CompositeAffine-3d-z0 | linear transforms | CompositeAffine in 3D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o r -z 0 -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t CompositeAffine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/CompositeAffine-3d-float | linear transforms | CompositeAffine in 3D, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t CompositeAffine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/CompositeAffine-2d | linear transforms | CompositeAffine in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t CompositeAffine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/CompositeAffine-2d-z0 | linear transforms | CompositeAffine in 2D, -z 0, -r [fixed,moving,1]command lineantsRegistration -d 2 -v 1 -o r -z 0 -r '[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -t CompositeAffine[0.1] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,0,Regular,0.5]' -c [10,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
transforms/CompositeAffine-2d-float | linear transforms | CompositeAffine in 2D, --float 1command lineantsRegistration -d 2 -v 1 -o r --float 1 -t CompositeAffine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Random,0.5]' -c [10,1e-6,5] -f 1 -s 1 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
transforms/CompositeAffine-4d | linear transforms | CompositeAffine in 4D (an affine transform in 4D)command lineantsRegistration -d 4 -v 1 -o r -t CompositeAffine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
transforms/two-stages | linear transforms | Rigid then Affinecommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
transforms/two-stages-z0 | linear transforms | Rigid then Affine, not collapsedcommand lineantsRegistration -d 3 -v 1 -o r -z 0 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in r1Affine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
transforms/oblique | linear transforms | Affine between oblique gridscommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
metrics/mattes-mi-alias | image metrics | Rigid with MI (mattes-mi-alias)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'MI[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/mi2 | image metrics | Rigid with MI2 (mi2)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 3.47e-18 in a parameter | 12/12 |
metrics/mi2-dense | image metrics | Rigid with MI2 (mi2-dense)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,20]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/meansquares | image metrics | Rigid with MeanSquares (meansquares)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/gc | image metrics | Rigid with GC (gc)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/gc-dense | image metrics | Rigid with GC (gc-dense)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1,0]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/cc-sampled | image metrics | Rigid with CC (cc-sampled)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/cc-dense | image metrics | Rigid with CC (cc-dense)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,3]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/mattes-gradient-filter | image metrics | Rigid with Mattes (mattes-gradient-filter)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25,1]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/cc-gradient-filter | image metrics | Rigid with CC (cc-gradient-filter)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2,None,1,1]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
metrics/mattes-gradient-filter-oblique | image metrics | Rigid with Mattes and the gradient filter, oblique gridscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1,32,Regular,0.25,1]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/cc-gradient-filter-oblique | image metrics | Rigid with CC and the gradient filter, oblique gridscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'CC[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1,2,None,1,1]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/mi2-2d | image metrics | MI2 in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Affine[0.1] -m 'MI2[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,24,Regular,0.5]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/cc-2d | image metrics | CC in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Affine[0.1] -m 'CC[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,4,None]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/gc-2d | image metrics | GC in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Affine[0.1] -m 'GC[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,0,Regular,0.5]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/mi2-4d-gradient-filter | image metrics | MI2 in 4Dcommand lineantsRegistration -d 4 -v 1 -o r -t Affine[0.1] -m 'MI2[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,20,Regular,0.5,1]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/mi2-float | image metrics | MI2, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t Rigid[0.1] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,32,Random,0.3]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/cc-float | image metrics | CC, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t Rigid[0.1] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2,Random,0.3]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/gc-float | image metrics | GC, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t Rigid[0.1] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1,0,Random,0.3]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/multi | image metrics | two metrics in one stage (weights 1 and 0.5)command lineantsRegistration -d 3 -v 1 -o r -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,0.5,0,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/demons | image metrics | Demons in a linear stage: ITK throws and ANTs drops the stage (ANTS-49)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Demons[<generated: reg-fixed>,<generated: reg-moving>,1,0,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
metrics/no-overlap | image metrics | no sample maps inside the moving image (Mattes throws, ANTS-49)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-far>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/random | metric sampling | Random 0.4command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Random,0.4]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/regular-1 | metric sampling | Regular 1 (every voxel, perturbed)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,1]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/sigmas-mm | metric sampling | smoothing sigmas in mmcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 2x1mm | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/shrink-4 | metric sampling | three levels, shrink 4x2x1command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [8x6x4,1e-6,5] -f 4x2x1 -s 2x1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat, standard output; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
sampling/winsorize-histogram | metric sampling | -w [0.005,0.995] -u 1command lineantsRegistration -d 3 -v 1 -o r -w [0.005,0.995] -u 1 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/masks | metric sampling | -x [mask,mask]command lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/fixed-mask-only | metric sampling | -x mask (fixed only), densecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: reg-mask>' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/moving-mask-null | metric sampling | -x [NULL,mask]command lineantsRegistration -d 3 -v 1 -o r -x '[NULL,<generated: reg-mask>]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
sampling/restrict-deformation | metric sampling | -g 1x1x0x1x1x1: a rotation frozencommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox -g 1x1x0x1x1x1 | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat, standard output; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
sampling/zero-iterations | metric sampling | a first level of 0 iterations runs nonecommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [0x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
sampling/no-seed | metric sampling | no seed: ANTs seeds its sampling from the clockcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Expected divergence without a seed ANTs seeds its sampling from the clock and its result changes from run to run; the replica uses a fixed seed | 1/1 |
initial/geometry | initial transforms | -r [fixed,moving,0]command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r '[<generated: reg-fixed>,<generated: reg-moving>,0]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
initial/moments | initial transforms | -r [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
initial/origins | initial transforms | -r [fixed,moving,2]command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r '[<generated: reg-fixed>,<generated: reg-moving>,2]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
initial/fixed-initial | initial transforms | -q [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o r -q '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
initial/file | initial transforms | -r a transform filecommand lineantsRegistration -d 3 -v 1 -o r -r initial.mat -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
initial/file-inverted | initial transforms | -r [file,1]command lineantsRegistration -d 3 -v 1 -o r -r [initial.mat,1] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
initial/per-stage | initial transforms | --initialize-transforms-per-stage 1: each stage starts from the previouscommand lineantsRegistration -d 3 -v 1 -o r -z 0 -a 1 -i 1 -t Translation[0.5] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [6,1e-6,5] -f 1 -s 0 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [6,1e-6,5] -f 1 -s 0 -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [6,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
initial/per-stage-from-r | initial transforms | -i 1 with -r [fixed,moving,1]command lineantsRegistration -d 3 -v 1 -o r -z 0 -a 1 -i 1 -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [6,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
initial/restore-state | initial transforms | --restore-state from a transform filecommand lineantsRegistration -d 3 -v 1 -o r -k initial.mat -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
initial/save-state | initial transforms | --save-state after linear stages: ANTs has no state and exits 1 (ANTS-50)command lineantsRegistration -d 3 -v 1 -o r -a 1 --save-state state.h5 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Rejected by both 1 | 2/2 |
initial/names-restored | initial transforms | -z 0 with a restored state: a file named after the wrong stage, then a read past the end of the names (ANTS-51)command lineantsRegistration -d 3 -v 1 -o r -z 0 -k initial.mat -t Affine[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Expected divergence ANTs reads past the end of its list of file names and crashes (ANTS-51); the replica writes the same files up to there and stops with an error | 1/1 |
dimensions-2d/warped | 2D images | the warped and inverse-warped images (linear)command lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/warped-nearest | 2D images | -n NearestNeighborcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n NearestNeighbor -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/warped-float | 2D images | Similarity, --float 1command lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] --float 1 -t Similarity[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/fixed-initial | 2D images | -q [fixed,moving,1]command lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -q '[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/file-affine | 2D images | -r an AffineTransform filecommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r affine-2d.mat -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/file-affine-inverted | 2D images | -r [AffineTransform file,1]command lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r [affine-2d.mat,1] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/file-euler | 2D images | -r a text Euler2DTransform file, -z 0command lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -r euler-2d.txt -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/file-euler-inverted | 2D images | -r [Euler2DTransform,1]: its own inverse classcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r [euler-2d.txt,1] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/file-rigid2d | 2D images | -r a Rigid2DTransform file, -z 0 -a 1command lineantsRegistration -d 2 -v 1 -o r -z 0 -a 1 -r rigid-2d.mat -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
dimensions-2d/file-similarity-inverted | 2D images | -r [Similarity2DTransform,1], --float 1command lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] --float 1 -r [similarity-2d.mat,1] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/file-translation | 2D images | -r a TranslationTransform filecommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r translation-2d.txt -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/file-unregistered | 2D images | -r an ANTSCenteredAffine2DTransform file: ITK's factory does not know the classcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r centered-affine-2d.mat -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both “VersorTransform_float_3_3” | 2/2 |
dimensions-2d/file-3d | 2D images | -r a 3D transform in a 2D registrationcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r initial.mat -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
dimensions-2d/composite | 2D images | -a 1 with -r [fixed,moving,1]command lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -a 1 -r '[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
dimensions-2d/interval-volumes | 2D images | --write-interval-volumes 2command lineantsRegistration -d 2 -v 1 -o r --write-interval-volumes 2 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
dimensions-2d/interpolation-BSpline | 2D images | -n BSpline[3] in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n BSpline[3] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/interpolation-BSpline5-float | 2D images | -n BSpline[5] in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n BSpline[5] --float 1 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/interpolation-Gaussian | 2D images | -n Gaussian in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n Gaussian -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-2d/interpolation-Gaussian-sigmas | 2D images | -n Gaussian[1.5x0.8,1.2] in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n Gaussian[1.5x0.8,1.2] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-2d/interpolation-LanczosWindowedSinc | 2D images | -n LanczosWindowedSinc in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n LanczosWindowedSinc -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-2d/interpolation-BlackmanWindowedSinc-float | 2D images | -n BlackmanWindowedSinc in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n BlackmanWindowedSinc --float 1 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/interpolation-WelchWindowedSinc | 2D images | -n WelchWindowedSinc in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n WelchWindowedSinc -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-2d/interpolation-GenericLabel | 2D images | -n GenericLabel in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n GenericLabel -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-2d/interpolation-MultiLabel | 2D images | -n MultiLabel[1.2] in 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n MultiLabel[1.2] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-4d/warped | 4D images | the warped and inverse-warped images (linear)command lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, r0GenericAffine.mat, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-4d/warped-nearest-float | 4D images | -n NearestNeighbor, --float 1command lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n NearestNeighbor --float 1 -t Translation[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-4d/file-affine-inverted | 4D images | -r [AffineTransform file,1]command lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r [affine-4d.mat,1] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-4d/interpolation-BSpline | 4D images | -n BSpline[2] in 4Dcommand lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n BSpline[2] -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-4d/interpolation-Gaussian-float | 4D images | -n Gaussian[1x1x1x1,1.5] in 4Dcommand lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n Gaussian[1x1x1x1,1.5] --float 1 -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-4d/interpolation-HammingWindowedSinc | 4D images | -n HammingWindowedSinc in 4Dcommand lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n HammingWindowedSinc -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-4d/interpolation-GenericLabel | 4D images | -n GenericLabel in 4Dcommand lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n GenericLabel -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
dimensions-4d/interpolation-MultiLabel-float | 4D images | -n MultiLabel in 4Dcommand lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n MultiLabel --float 1 -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [4,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
dimensions-4d/composite-z0 | 4D images | -a 1 -z 0 -r [fixed,moving,0]command lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz,iw.nii.gz] -a 1 -z 0 -r '[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,0]' -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
dimensions-4d/interval-volumes | 4D images | --write-interval-volumes 1command lineantsRegistration -d 4 -v 1 -o r --write-interval-volumes 1 -t Affine[0.1] -m 'Mattes[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1,32,Regular,0.25]' -c [3,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
outputs/interpolation-NearestNeighbor | outputs | warped images with -n NearestNeighborcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n NearestNeighbor -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
outputs/interpolation-BSpline | outputs | warped images with -n BSpline[3]command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n BSpline[3] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
outputs/interpolation-LanczosWindowedSinc | outputs | warped images with -n LanczosWindowedSinccommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n LanczosWindowedSinc -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
outputs/interpolation-MultiLabel | outputs | warped images with -n MultiLabelcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n MultiLabel -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
outputs/interpolation-Gaussian | outputs | warped images with -n Gaussiancommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -n Gaussian -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5,1e-6,5] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in iw.nii.gz, w.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
outputs/print-similarity | outputs | --print-similarity-measure-interval 2 (full-scale CC)command lineantsRegistration -d 3 -v 1 -o r --print-similarity-measure-interval 2 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
outputs/write-interval-volumes | outputs | --write-interval-volumes 3command lineantsRegistration -d 3 -v 1 -o r --write-interval-volumes 3 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [6x4,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 12/12 |
errors/help | errors | --help: the menu, exit 0command lineantsRegistration --help | Bit-identical exit status, standard output, standard error, every file | 5/5 |
errors/no-arguments | errors | no arguments: the short menucommand lineantsRegistration | Rejected by both 1 | 2/2 |
errors/bad-number | errors | -c [1x,1e-6,5]: a number that does not convertcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [1x,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
errors/levels-mismatch | errors | two shrink factors for one levelcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [5] -f 2x1 -s 0 | Rejected by both 1 | 2/2 |
errors/missing-image | errors | a metric image that does not exist: ANTs uses the null image and crashes (ANTS-02)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,none.nii.gz,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both file /tmp/lx-reg-anr9n5ej/none.nii.gz does not exist . | 0/2 |
errors/unknown-transform | errors | an unknown transform namecommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Foo[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both Unknown registration method “foo” | 2/2 |
errors/save-state-without-composite | errors | --save-state without -a 1command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] --save-state s.h5 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both because the the output transform will contain the this processes initializer | 2/2 |
errors/per-stage-with-collapse | errors | -i 1 -a 1 with collapsingcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -a 1 -i 1 -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both ERROR: initialize-transforms-per-stage & collapse-output-transforms options are mutually exclusive. | 2/2 |
errors/restore-with-r | errors | --restore-state with -rcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -k initial.mat -r initial.mat -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10x5,1e-6,5] -f 2x1 -s 1x0vox | Rejected by both restore-state option is mutually exclusive with initial-moving-transform & initial-fixed-transform options. | 2/2 |
errors/short-help | errors | -h: the short menu, exit 0command lineantsRegistration -h | Bit-identical exit status, standard output, standard error, every file | 5/5 |
errors/version | errors | --versioncommand lineantsRegistration --version | Expected divergence The replica prints its own version and the ANTs version it ports | 1/1 |
errors/version-0 | errors | --version 0: no version, then no dimensioncommand lineantsRegistration --version 0 | Expected divergence ANTs exits 1 without a word (the reason is printed only with --verbose, which this command line does not set); the replica prints the reason (ANTS-45) | 1/1 |
errors/no-dimension | errors | no -d, verbose: the dimension from the first fixed image is unknowncommand lineantsRegistration -o o -v | Rejected by both Image dimensionality not specified. See command line option --dimensionality | 2/2 |
errors/no-dimension-quiet | errors | no -d, quietcommand lineantsRegistration -o o | Expected divergence ANTs exits 1 without a word (the reason is printed only with --verbose, which this command line does not set); the replica prints the reason (ANTS-45) | 1/1 |
errors/bad-dimension | errors | -d 5command lineantsRegistration -d 5 -v | Rejected by both bad image dimension 5 | 2/2 |
errors/invalid-flag | errors | an option ANTs does not knowcommand lineantsRegistration -d 3 --nope 1 -y | Rejected by both ERROR: Invalid command line flags found! Aborting execution. | 2/2 |
errors/unbalanced-bracket | errors | -t Rigid[0.1 (no closing bracket): the rest of the line joins the transformcommand lineantsRegistration -d 3 -t Rigid[0.1 -v | Expected divergence ANTs exits 1 without a word (the reason is printed only with --verbose, which this command line does not set); the replica prints the reason (ANTS-45) | 1/1 |
errors/misplaced-bracket | errors | -t Rig]id[0.1]command lineantsRegistration -d 3 -t Rig]id[0.1] -v | Rejected by both ERROR: the metric option (‘-m’) must be specified. See help menu. | 2/2 |
errors/no-transform | errors | no -tcommand lineantsRegistration -d 3 -v -o o | Rejected by both ERROR: the transform option (‘-t’) must be specified. See help menu. | 2/2 |
errors/no-metric | errors | no -mcommand lineantsRegistration -d 3 -v -o o -t Rigid[0.1] | Rejected by both ERROR: the metric option (‘-m’) must be specified. See help menu. | 2/2 |
errors/no-convergence | errors | no -ccommand lineantsRegistration -d 3 -v -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' | Rejected by both ERROR: the convergence option (‘-c’) must be specified. See help menu. | 2/2 |
errors/no-shrink-factors | errors | no -fcommand lineantsRegistration -d 3 -v -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 | Rejected by both ERROR: the shrink factors option (‘-f’) must be specified. See help menu. | 2/2 |
errors/no-smoothing-sigmas | errors | no -scommand lineantsRegistration -d 3 -v -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 | Rejected by both ERROR: the smoothing sigmas option (‘-s’) must be specified. See help menu. | 2/2 |
errors/no-output | errors | no -ocommand lineantsRegistration -d 3 -v -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 -s 0 | Rejected by both ERROR: the output option (‘-o’) must be specified. See help menu. | 2/2 |
errors/per-stage-without-composite | errors | -i 1 without -a 1command lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 -s 0 -v -i 1 | Rejected by both because the initializizing transform is collapsed into each stage for optimization | 2/2 |
errors/convergence-count | errors | two -c for one stagecommand lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 -s 0 -v -c 10 | Rejected by both ERROR: smoothing sigmas, shrink factors, and convergence options must be set once for each transform. | 2/2 |
errors/window-1 | errors | -c [10,1e-6,1]: a convergence window of 1command lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -v -c [10,1e-6,1] -f 1 -s 0 | Rejected by both Convergence Window Size must be greater than or equal to 2 | 2/2 |
errors/bad-iterations | errors | -c [1ox,1e-6,5]command lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -v -c [1ox,1e-6,5] -f 1 -s 0 | Rejected by both ERROR: Unable to convert ‘1o’ to type ‘j’ as unsigned int | 2/2 |
errors/unknown-metric | errors | an unknown metriccommand lineantsRegistration -d 3 -o o -t Rigid[0.1] -v -m 'Foo[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 -s 0 | Rejected by both 1 | 2/2 |
errors/bad-sampling | errors | an unknown sampling strategycommand lineantsRegistration -d 3 -o o -t Rigid[0.1] -v -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Sparse,0.2]' -c 10 -f 1 -s 0 | Rejected by both ERROR: invalid sampling strategy specified: sparse | 2/2 |
errors/sigmas-mismatch | errors | three sigmas for two levelscommand lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -v -c 10x5 -f 2x1 -s 1x0x0 | Rejected by both 1 | 2/2 |
errors/stage-without-metric | errors | a second -t without its -mcommand lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 -s 0 -v -t Affine[0.1] -c 10 -f 1 -s 0 | Rejected by both The number of transforms is 2 and the last stage ID as determined by the metrics is 0. | 2/2 |
errors/sampling-above-1 | errors | a sampling percentage of 1.5command lineantsRegistration -d 3 -o o -t Rigid[0.1] -v -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,1.5]' -c 10 -f 1 -s 0 | Rejected by both Description: ITK ERROR: ImageRegistrationMethodv4(0x62cd28a33bc0): sampling percentage outside expected (0,1] range | 2/2 |
errors/missing-initial | errors | -r a transform file that does not existcommand lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 -s 0 -v -r none.mat | Rejected by both Transform file does not exist: /tmp/lx-reg-e4ldfnfe/none.mat | 2/2 |
errors/minc | errors | --minc 1: MINC transform filescommand lineantsRegistration -d 3 -o o -t Rigid[0.1] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,32]' -c 10 -f 1 -s 0 -v --minc 1 | Expected divergence ANTs 2.6.5 has no MINC transform IO (it prints ITK’s error and exits 0); the replica refuses MINC transform files with a message | 1/1 |
point-sets/icp | point-set metrics | ICP between two label images, the fixed mask as the virtual domaincommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
point-sets/icp-affine-two-levels | point-set metrics | ICP, Affine, two levels (shrink 2x1, smoothing ignored)command lineantsRegistration -d 3 -v 1 -o r -t Affine[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5x5 -f 2x1 -s 1x0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-float | point-set metrics | ICP with --float 1command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' --float 1 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-text | point-set metrics | ICP between text point sets (a final newline: ANTs reads one more record, ANTS-62)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m ICP[fixed.txt,moving.txt,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-text-no-final-newline | point-set metrics | A text point set without a final newlinecommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m ICP[fixed-no-final-newline.txt,moving.txt,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-vtk | point-set metrics | ICP between ASCII VTK point setscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m ICP[fixed.vtk,moving.vtk,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-vtk-binary | point-set metrics | A binary VTK point set: its labels read past one value (ANTS-63)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m ICP[fixed-binary.vtk,moving.vtk,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-vtk-no-scalars | point-set metrics | A VTK point set without labels: label 0 for every pointcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m ICP[fixed-no-scalars.vtk,moving.vtk,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-boundary | point-set metrics | ICP on the labels' contours (boundaryPointsOnly)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-not-bijective | point-set metrics | A label only the fixed set has: ITK throws, the stage is dropped (ITK-22)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving-partial>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-empty | point-set metrics | An empty fixed point set: the scales estimator throwscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-zero>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-sampled | point-set metrics | ICP on half of each point set (a sampling percentage of 0.5)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,0.5]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Expected divergence ANTs draws the points of a sampling percentage below 1 with a generator seeded from the clock (ANTS-64): two runs differ; the replica seeds it from the registration’s seed | 1/1 |
point-sets/icp-2d | point-set metrics | ICP in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed-2d>,<generated: ps-labels-moving-2d>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask-2d>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-4d | point-set metrics | ICP in 4D, Affinecommand lineantsRegistration -d 4 -v 1 -o r -t Affine[0.1] -m 'ICP[<generated: ps-labels-fixed-4d>,<generated: ps-labels-moving-4d>,1]' -c 3 -f 1 -s 0 -x '<generated: ps-mask-4d>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-fixed-initial | point-set metrics | ICP with an initial fixed transform (-q)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' -q initial.mat | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-moving-initial | point-set metrics | ICP with an initial moving transform (-r)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' -r initial.mat | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-images-initial | point-set metrics | ICP with -r [fixed,moving,1] from intensity imagescommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' -r '[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1]' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-same-sets | point-set metrics | ICP of a point set with itselfcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-fixed>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/pse | point-set metrics | PSE (sigma 1, 50 neighbours)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'PSE[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,50]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
point-sets/pse-boundary | point-set metrics | PSE on the contours, 20 neighbourscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'PSE[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,1,1,20]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/pse-sigma-zero | point-set metrics | PSE with a point-set sigma of 0: ITK throwscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'PSE[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,0,50]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/pse-text-2d | point-set metrics | PSE between 2D text point setscommand lineantsRegistration -d 2 -v 1 -o r -t Rigid[0.1] -m PSE[fixed-2d.txt,moving-2d.txt,1,1,0,1,10] -c 5 -f 1 -s 0 -x '<generated: ps-mask-2d>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/jhct | point-set metrics | JHCT with anisotropic covariances (sigma 1, alpha 1.1)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,50,1.1,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
point-sets/jhct-isotropic | point-set metrics | JHCT with isotropic covariances (sigma 1: ITK's univariate prefactor, ITK-23)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,50,1.1,0]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/jhct-alpha-1 | point-set metrics | JHCT with alpha 1 (the logarithm)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,50,1,0]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/jhct-alpha-2 | point-set metrics | JHCT with alpha 2command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,50,2,0]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/jhct-sigma-2 | point-set metrics | JHCT with a point-set sigma of 2command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,2,50,1.1,0]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/jhct-anisotropic-sigma-2 | point-set metrics | JHCT, anisotropic, sigma 2, alpha 1.5command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,2,50,1.5,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/jhct-text | point-set metrics | JHCT between text point sets (fewer points than neighbours: ITK's warnings)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m JHCT[fixed.txt,moving.txt,1,1,0,1,50,1.1,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/jhct-2d | point-set metrics | JHCT in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed-2d>,<generated: ps-labels-moving-2d>,1,1,0,1,20,1.1,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask-2d>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/igdm | point-set metrics | IGDM between masked intensity images (sigmas estimated)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
point-sets/igdm-sigmas | point-set metrics | IGDM with given sigmascommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>,0x0x0,50,5,1,2]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/igdm-fixed-initial | point-set metrics | IGDM with an initial fixed transformcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' -q initial.mat | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/igdm-neighbourhood | point-set metrics | IGDM with a neighbourhood of radius 1command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>,1x1x1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Expected divergence ANTs’ line search computes IGDM’s centre index from a negative number and reads memory before the intensity arrays (ANTS-66, undefined); the replica takes the two intensities as equal | 1/1 |
point-sets/igdm-2d | point-set metrics | IGDM in 2D, radius 1command lineantsRegistration -d 2 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed-2d>,<generated: ps-intensity-moving-2d>,1,<generated: ps-mask-2d>,<generated: ps-mask-moving-2d>,1x1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask-2d>' | Expected divergence ANTs reads memory before IGDM’s intensity arrays for neighbourhoods of more than one voxel (ANTS-66) | 1/1 |
point-sets/igdm-2d-radius-0 | point-set metrics | IGDM in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed-2d>,<generated: ps-intensity-moving-2d>,1,<generated: ps-mask-2d>,<generated: ps-mask-moving-2d>]' -c 5 -f 1 -s 0 -x '<generated: ps-mask-2d>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/mattes-then-icp | point-set metrics | Mattes and ICP in one stage (the image metric's virtual domain)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,32]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-then-mattes | point-set metrics | ICP and Mattes in one stage, weights 0.5 and 2command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,0.5]' -m 'Mattes[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,2,32]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-then-mattes-regular | point-set metrics | ICP first: the stage takes ICP's sampling (none)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -m 'Mattes[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,32,Regular,0.5]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/igdm-then-mattes | point-set metrics | IGDM and Mattes in one stagecommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -m 'Mattes[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,32]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-then-igdm | point-set metrics | ICP and IGDM in one stage: IGDM gets no input, ITK throws (ANTS-61)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/mattes-then-igdm | point-set metrics | Mattes and IGDM in one stage: IGDM gets no input (ANTS-61)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,32]' -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/icp-then-pse-stages | point-set metrics | Two stages: ICP (Rigid), then PSE (Affine)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 3 -f 1 -s 0 -t Affine[0.1] -m 'PSE[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,20]' -c 3 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
point-sets/mattes-and-icp-warped | point-set metrics | Mattes first, then ICP: the warped images of the image metriccommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Rigid[0.1] -m 'Mattes[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,32]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Bit-identical exit status, standard output, standard error, every file | 8/8 |
point-sets/igdm-few-parameters | point-set metrics | IGDM with three parameterscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both The expected number of parameters aren’t specified. Please see help menu. | 2/2 |
point-sets/igdm-radius-dimension | point-set metrics | IGDM with a 2D radius in 3Dcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>,1x1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both The neighborhood size must equal the dimension. | 2/2 |
point-sets/icp-percentage-above-1 | point-set metrics | A sampling percentage of 1.5command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1.5]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both Description: ITK ERROR: ImageRegistrationMethodv4(0x5a2048a17e60): sampling percentage outside expected (0,1] range | 2/2 |
point-sets/icp-no-mask | point-set metrics | Point-set metrics without a fixed mask: no virtual domaincommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 | Rejected by both larmorx: antsRegistration: a stage of point-set metrics needs a fixed mask (-x) to define its virtual domain: ANTs reads the spacing of a null image; ANTs crashes here, the replica stops | 4/4 |
point-sets/icp-missing | point-set metrics | A point set that does not existcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[none.nii.gz,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both larmorx: antsRegistration: a labelled point set could not be read: ANTs goes on with a null point set and crashes | 4/4 |
point-sets/icp-short-name | point-set metrics | A point set named with fewer than 3 characterscommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[ab,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both larmorx: antsRegistration: a labelled point set could not be read: ANTs goes on with a null point set and crashes | 4/4 |
point-sets/icp-unreadable | point-set metrics | A point set no reader takes (ITK's image reader throws)command lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m ICP[points.csv,moving.txt,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both larmorx: upstream defect ANTS-62 (antsRegistration) reproduced: ‘/tmp/lx-reg-4d3i0_bn/moving.txt’ ends with a newline: an extra point, a copy of the last one, as the oracle’s build adds it from uninit | 4/4 |
point-sets/icp-vtk-no-points | point-set metrics | A VTK file without POINTS: libstdc++'s out_of_rangecommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m ICP[no-points.vtk,moving.vtk,1] -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both larmorx: antsRegistration: ANTs aborts here (an uncaught exception); the replica stops with exit 1 | 4/4 |
point-sets/igdm-missing-mask | point-set metrics | An IGDM mask that does not existcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,none.nii.gz,<generated: ps-mask-moving>]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both larmorx: upstream defect ANTS-68 (antsRegistration) reproduced: IGDM’s derivative lacks the 1/sigma^2 factors of its two terms, as in ANTs [LARMORX_UPSTREAM_DEFECTS=fix corrects it] | 4/4 |
point-sets/icp-warped | point-set metrics | Warped images with a point-set metric firstcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both larmorx: antsRegistration: warped images with a point-set metric first: ANTs resamples null images and crashes, the replica stops | 4/4 |
point-sets/icp-print-similarity | point-set metrics | --print-similarity-measure-interval with a point-set metric firstcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' --print-similarity-measure-interval 1 | Rejected by both larmorx: antsRegistration: --print-similarity-measure-interval with a point-set metric first: ANTs’ observer computes the full-scale similarity of null images; ANTs crashes here, the replica stops | 4/4 |
point-sets/icp-interval-volumes | point-set metrics | --write-interval-volumes with a point-set metric firstcommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' --write-interval-volumes 1 | Rejected by both larmorx: antsRegistration: --write-interval-volumes with a point-set metric first: ANTs’ observer resamples a null image; ANTs crashes here, the replica stops | 4/4 |
point-sets/mattes-regular-then-icp | point-set metrics | Regular sampling with a point-set metric in the stagecommand lineantsRegistration -d 3 -v 1 -o r -t Rigid[0.1] -m 'Mattes[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,32,Regular,0.5]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c 5 -f 1 -s 0 -x '<generated: ps-mask>' | Rejected by both larmorx: antsRegistration: a stage that samples its metrics (Regular or Random) with a point-set metric: ANTs gives the sample points to a null image metric and crashes; ANTs crashes here, the replica sto | 4/4 |
syn/cc | SyN | SyN with CC (radius 2), 2 levels, warped imagescommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 10/10 |
syn/meansquares-total | SyN | SyN with MeanSquares, a total field variance of 0.5command lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 8/8 |
syn/mattes | SyN | SyN with Mattes (24 bins), every voxelcommand lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 2.13e-14 in a parameter | 14/14 |
syn/mattes-regular | SyN | SyN with Mattes, regular sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24,Regular,0.5]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
syn/mattes-random | SyN | SyN with Mattes, random sampling 0.6command lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24,Random,0.6]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/mi2 | SyN | SyN with the joint-histogram MI (MI2, ITK-18 reproduced)command lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,20]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
syn/demons | SyN | SyN with Demons (the fixed image's gradient)command lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0] -m 'Demons[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/gc | SyN | SyN with GC: ITK adds every point's derivative to the first voxel (ITK-30)command lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/multi | SyN | two metrics (CC 0.7, Mattes 0.3): the weights normalised at each evaluationcommand lineantsRegistration -d 3 -v 1 -o r -t SyN[0.25,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,0.7,2]' -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.3,20]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/masks | SyN | -x [mask,mask]: both masks resampled through the fieldscommand lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/fixed-mask-only | SyN | a fixed mask on another grid, no moving mask (ITK-31)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: reg-mask-oblique>' -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/moving-mask-only | SyN | -x [NULL,mask] with the mask on another grid (ITK-31)command lineantsRegistration -d 3 -v 1 -o r -x '[NULL,<generated: reg-mask-oblique>]' -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/rigid-affine-syn | SyN | Rigid, Affine then SyN, collapsed: 0GenericAffine.mat, 1Warp, 1InverseWarpcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t Rigid[0.1] -m 'MI[<generated: reg-fixed>,<generated: reg-moving>,1,32,Regular,0.25]' -c [10,1e-6,5] -f 1 -s 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
syn/no-collapse | SyN | Affine then SyN, -z 0command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
syn/composite | SyN | Affine then SyN, -a 1: the composite and its inverse hold the fieldscommand lineantsRegistration -d 3 -v 1 -o r -a 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/composite-no-collapse | SyN | Affine then SyN, -a 1 -z 0command lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/two-stages | SyN | two SyN stages, collapsed into one fieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox -t SyN[0.15,2,0.3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/two-stages-no-collapse | SyN | two SyN stages, -z 0command lineantsRegistration -d 3 -v 1 -o r -z 0 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox -t SyN[0.15,2,0.3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/save-state | SyN | --save-state after SyN: the fixed-to-middle and moving-to-middle fields and inversescommand lineantsRegistration -d 3 -v 1 -o r -a 1 --save-state state.h5 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
syn/save-state-then-translation | SyN | --save-state after SyN then Translation: ANTs saves the SyN state without the translation (ANTS-74)command lineantsRegistration -d 3 -v 1 -o r -a 1 --save-state state.h5 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox -t Translation[0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4,1e-6,3] -f 1 -s 0 | Expected divergence a linear stage after a displacement field: its learning rate comes from voxels ANTs draws with a generator seeded from the clock (ANTS-72); the saved state, the last SyN stage’s, is identical to ANTs’ (checked by hand, 2026-10-19) | 1/1 |
syn/restore-state-then-translation | SyN | --restore-state of a SyN state, then Translation, --save-state (ANTS-74)command lineantsRegistration -d 3 -v 1 -o r -a 1 -k state-3d.h5 --save-state state.h5 -t Translation[0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4,1e-6,3] -f 1 -s 0 | Expected divergence a linear stage after a displacement field: its learning rate comes from voxels ANTs draws with a generator seeded from the clock (ANTS-72); the saved state, the last SyN stage’s, is identical to ANTs’ (checked by hand, 2026-10-19) | 1/1 |
syn/restore-state | SyN | --restore-state of a SyN state: the composite ends with SyN's transformcommand lineantsRegistration -d 3 -v 1 -o r -a 1 -k state-3d.h5 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/restore-state-per-stage | SyN | --restore-state with -i 1: SyN starts from the restored fieldscommand lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -i 1 -k state-3d.h5 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/per-stage | SyN | -i 1: the second SyN stage starts from the first one's fieldscommand lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -i 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox -t SyN[0.15,2,0.3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/interval-volumes | SyN | --write-interval-volumes 2command lineantsRegistration -d 3 -v 1 -o r --write-interval-volumes 2 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 13/13 |
syn/print-similarity | SyN | --print-similarity-measure-interval 2 with initial transforms on both sidescommand lineantsRegistration -d 3 -v 1 -o r -r identity -q identity --print-similarity-measure-interval 2 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
syn/print-similarity-crash | SyN | --print-similarity-measure-interval without -q: a null transform (ANTS-71)command lineantsRegistration -d 3 -v 1 -o r --print-similarity-measure-interval 2 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both larmorx: upstream defect ANTS-71 (antsRegistration) reproduced: stage 0: ANTs crashes computing the full-scale similarity; the replica stops there [LARMORX_UPSTREAM_DEFECTS=fix corrects it] | 4/4 |
syn/print-similarity-multi-crash | SyN | --print-similarity-measure-interval with two metrics: a null metric (ANTS-71)command lineantsRegistration -d 3 -v 1 -o r -r identity -q identity --print-similarity-measure-interval 2 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both larmorx: upstream defect ANTS-71 (antsRegistration) reproduced: stage 0: ANTs crashes computing the full-scale similarity; the replica stops there [LARMORX_UPSTREAM_DEFECTS=fix corrects it] | 4/4 |
syn/restrict-deformation | SyN | -g 1x0.5x0: the update weighted per axiscommand lineantsRegistration -d 3 -v 1 -o r -g 1x0.5x0 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/2d | SyN | SyN in 2D, three levelscommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t SyN[0.3,2,0.5] -m 'CC[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,3]' -c [5x4x3,1e-7,3] -f 4x2x1 -s 2x1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/2d-float | SyN | SyN in 2D, --float 1command lineantsRegistration -d 2 -v 1 -o r --float 1 -t SyN[0.3,2,0] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -c [4x3,1e-7,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/4d | SyN | SyN in 4Dcommand lineantsRegistration -d 4 -v 1 -o [r,w.nii.gz] -t SyN[0.3,2,0] -m 'MeanSquares[<generated: reg-fixed-4d>,<generated: reg-moving-4d>,1]' -c [3x2,1e-7,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
syn/float | SyN | SyN, --float 1command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] --float 1 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/float-mattes-mask | SyN | SyN with Mattes and a fixed mask, --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -x '<generated: reg-mask>' -t SyN[0.25,3,0] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz, standard output; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
syn/gradient-filter | SyN | CC and Demons with the gradient filtercommand lineantsRegistration -d 3 -v 1 -o r -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2,None,1,1]' -m 'Demons[<generated: reg-fixed>,<generated: reg-moving>,1,0,None,1,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/mm-winsorize-histogram | SyN | sigmas in mm, -w and -u, an update variance of 0command lineantsRegistration -d 3 -v 1 -o r -w [0.01,0.99] -u 1 -t SyN[0.2,0,1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 2x0mm | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/sdcflows-syn-sdc | SyN | sdcflows' fieldmap-less SyN (sd_syn.json): two SyN stages of two Mattes metrics (random sampling, then dense), -g along the phase-encoding axis, masks, histogram matching, winsorizingcommand lineantsRegistration -d 3 -v 1 -o r -w [0.001,0.998] -u 1 -x '[<generated: reg-mask>,<generated: reg-mask>]' -g 0.1x1x0.1 -t SyN[0.8,6,10] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.9,48,Random,0.8]' -m 'Mattes[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,0.1,48,Random,0.8]' -c [200x100,1e-6,5] -f 1x1 -s 2x0vox -g 0.1x1x0.1 -t SyN[0.8,2,0.5] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.8,48]' -m 'Mattes[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,0.2,48]' -c [10,1e-8,2] -f 1 -s 0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz, standard output; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
syn/fixed-initial | SyN | -q [fixed,moving,1]: the fixed images resampled through itcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -q '[<generated: reg-fixed>,<generated: reg-moving>,1]' -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/convergence | SyN | a level stopped by the convergence test, a level of 0 iterationscommand lineantsRegistration -d 3 -v 1 -o r -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [20x0x5,1e-1,4] -f 4x2x1 -s 2x1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/oblique | SyN | SyN between oblique gridscommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/default-variances | SyN | SyN[0.2]: the variances default to 3 and 0command lineantsRegistration -d 3 -v 1 -o r -t SyN[0.2] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn/quiet | SyN | Affine then SyN without -vcommand lineantsRegistration -d 3 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
syn/then-linear | SyN | SyN then Affine: the linear stage's scales from random voxels (ANTS-72)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox | Expected divergence a linear stage after a displacement field estimates its parameter scales from voxels drawn with a generator ANTs seeds from the clock (ANTS-72): two runs of ANTs differ; the replica seeds it from the registration’s seed | 1/1 |
syn/initial-field | SyN | -r a displacement field imagecommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r warp-3d.nii.gz -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/initial-field-linear | SyN | -r a displacement field image, then Affine (ANTS-72)command lineantsRegistration -d 3 -v 1 -o r -r warp-3d.nii.gz -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox | Expected divergence a linear stage after a displacement field estimates its parameter scales from voxels drawn with a generator ANTs seeds from the clock (ANTS-72) | 1/1 |
syn/initial-field-inverse | SyN | -r [field,1]: a field without an inversecommand lineantsRegistration -d 3 -v 1 -o r -r [warp-3d.nii.gz,1] -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
syn/initial-composite | SyN | -r a composite holding an affine transform and a fieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r composite-3d.h5 -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
syn/initial-composite-inverse | SyN | -r [composite,1]: its field has no inversecommand lineantsRegistration -d 3 -v 1 -o r -r [composite-3d.h5,1] -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
syn/fixed-field | SyN | -q a displacement field imagecommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -q warp-3d.nii.gz -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/initial-field-2d | SyN | -r a 2D displacement field imagecommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r warp-2d.nii.gz -t SyN[0.3,2,0] -m 'CC[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
syn/initial-scalar | SyN | -r a scalar image: ITK reads it as a field of equal components (ANTS-73)command lineantsRegistration -d 3 -v 1 -o r -r '<generated: reg-mask>' -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
syn/initial-scalar-2d | SyN | -r a 2D scalar image: ITK reads it as complex pixels, both parts equal (ANTS-73)command lineantsRegistration -d 2 -v 1 -o r -r '<generated: reg-mask-2d>' -t SyN[0.3,2,0] -m 'CC[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
syn/initial-field-2d-in-3d | SyN | -r a 2D field in 3D: one slice, each vector the product of its components (ANTS-73)command lineantsRegistration -d 3 -v 1 -o r -r warp-2d.nii.gz -t SyN[0.2,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
syn/igdm | SyN | IGDM first in a SyN stage: not implemented in ANTscommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: reg-mask>' -t SyN[0.2,3,0] -m 'IGDM[<generated: reg-fixed>,<generated: reg-moving>,1,<generated: reg-mask>,<generated: reg-mask>]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
syn/bad-variance | SyN | SyN[0.2,3x]: a variance that does not convertcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t SyN[0.2,3x,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both ERROR: Unable to convert ‘3x’ to type ‘f’ as float | 2/2 |
syn/exponential | SyN | Exponential[0.1,3,0] with CC in SyN's settings (the velocity-field stages have their own categories)command lineantsRegistration -d 3 -v 1 -o r -t Exponential[0.1,3,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn-point-sets/icp | SyN with point sets | ICP alone: the fixed mask's grid as the virtual domaincommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 8/8 |
syn-point-sets/pse | SyN with point sets | PSE (sigma 1, 20 neighbours)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'PSE[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,20]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn-point-sets/jhct | SyN with point sets | JHCT, anisotropic (sigma 1, alpha 1.1)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,20,1.1,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
syn-point-sets/icp-float | SyN with point sets | ICP, --float 1command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' --float 1 -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn-point-sets/icp-moving-initial | SyN with point sets | ICP after -r (a matrix)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -r initial.mat -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
syn-point-sets/icp-fixed-initial | SyN with point sets | ICP after -q (a matrix)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -q initial.mat -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn-point-sets/icp-not-bijective | SyN with point sets | a fixed label the moving set lacks: ITK throws at the first such point (ITK-22)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving-partial>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both LabeledPointSetToPointSetMetricv4 (0x63b1c4fb59d0): The label sets are not bijective. | 2/2 |
syn-point-sets/icp-moving-extra-label | SyN with point sets | a moving label the fixed set lacks: ITK throws in the moving-to-middle update (ITK-22)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-moving-partial>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both LabeledPointSetToPointSetMetricv4 (0x62024e39e3b0): The label sets are not bijective. | 2/2 |
syn-point-sets/cc-and-icp | SyN with point sets | CC and ICP: ITK gives ICP identity transforms (ITK-33)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'CC[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,2]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,0.5]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn-point-sets/icp-and-cc | SyN with point sets | ICP first, then CC: ICP's virtual domain (ITK-33)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,0.5]' -m 'CC[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn-point-sets/cc-and-igdm | SyN with point sets | CC and IGDM: SyN holds labelled point sets only, IGDM gets no input (ANTS-61)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'CC[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,2]' -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
syn-point-sets/cc-igdm-icp | SyN with point sets | CC, IGDM and ICP: IGDM without point sets before a metric with them (ANTS-61)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'CC[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,2]' -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both larmorx: upstream defect ANTS-68 (antsRegistration) reproduced: IGDM’s derivative lacks the 1/sigma^2 factors of its two terms, as in ANTs [LARMORX_UPSTREAM_DEFECTS=fix corrects it] | 4/4 |
syn-point-sets/icp-2d | SyN with point sets | ICP in 2Dcommand lineantsRegistration -d 2 -v 1 -o r -x '<generated: ps-mask-2d>' -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed-2d>,<generated: ps-labels-moving-2d>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
syn-point-sets/cc-regular-and-icp | SyN with point sets | CC sampled (Regular) and ICP: ANTs gives ICP's sample points to a null image metriccommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t SyN[0.3,2,0] -m 'CC[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,2,Regular,0.5]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both larmorx: antsRegistration: a stage that samples its metrics (Regular or Random) with a point-set metric: ANTs gives the sample points to a null image metric and crashes; ANTs crashes here, the replica sto | 4/4 |
syn-point-sets/icp-field-initial | SyN with point sets | ICP after -r a displacement field without an inverse: ITK maps the moving points with a null transformcommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -r warp-3d.nii.gz -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both larmorx: antsRegistration: the moving transform of a point-set metric in a SyN stage cannot be inverted (a linear transform with a singular matrix, or a displacement field without an inverse): ITK map | 4/4 |
syn-point-sets/icp-no-mask | SyN with point sets | ICP alone without a fixed mask: no virtual domaincommand lineantsRegistration -d 3 -v 1 -o r -t SyN[0.3,2,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both larmorx: antsRegistration: a stage of point-set metrics needs a fixed mask (-x) to define its virtual domain: ANTs reads the spacing of a null image; ANTs crashes here, the replica stops | 4/4 |
bspline-syn/cc | B-spline SyN | BSplineSyN (mesh 2x2x2) with CC (radius 2), 2 levels, warped imagescommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t BSplineSyN[0.2,2x2x2,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 10/10 |
bspline-syn/knot-spacing | B-spline SyN | BSplineSyN[0.1,26,0,3] (antsRegistrationSyN.sh's): a knot spacing of 26 mmcommand lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.1,26,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 8/8 |
bspline-syn/total-mesh | B-spline SyN | a total-field mesh (3x3x3, update 4x4x4)command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,4x4x4,3x3x3,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/order-2 | B-spline SyN | B-splines of order 2command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,3x3x3,0,2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/order-5 | B-spline SyN | B-splines of order 5command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/mesh-0 | B-spline SyN | an update mesh of 0: the update field is not smoothedcommand lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,0x0x0,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/mattes | B-spline SyN | Mattes (24 bins)command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,3] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz, standard output; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
bspline-syn/mi2 | B-spline SyN | the joint-histogram MI (MI2)command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,3] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,20]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/gc | B-spline SyN | GC: ITK adds every point's derivative to the first voxel (ITK-30)command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,3] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/demons | B-spline SyN | Demonscommand lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,3] -m 'Demons[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/multi | B-spline SyN | CC (0.7) and Mattes (0.3)command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,0.7,2]' -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.3,20]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/masks | B-spline SyN | -x [mask,mask]: the B-spline fits weighted by the masks resampled through the fieldscommand lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t BSplineSyN[0.25,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/fixed-mask-only | B-spline SyN | a fixed mask on another grid, no moving maskcommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: reg-mask-oblique>' -t BSplineSyN[0.25,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/regular | B-spline SyN | MeanSquares, regular sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,1,Regular,0.5]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/random | B-spline SyN | Mattes, random sampling 0.6command lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,3] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24,Random,0.6]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/float | B-spline SyN | --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t BSplineSyN[0.2,2x2x2,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/2d | B-spline SyN | 2D, CCcommand lineantsRegistration -d 2 -v 1 -o r -t BSplineSyN[0.3,3x3,0,3] -m 'CC[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/affine-then | B-spline SyN | Affine then BSplineSyN, collapsed: 0GenericAffine.mat, 1Warp, 1InverseWarpcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t BSplineSyN[0.2,2x2x2,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
bspline-syn/composite | B-spline SyN | Affine then BSplineSyN, -a 1command lineantsRegistration -d 3 -v 1 -o r -a 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t BSplineSyN[0.2,2x2x2,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/no-collapse | B-spline SyN | Affine then BSplineSyN, -z 0command lineantsRegistration -d 3 -v 1 -o r -z 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t BSplineSyN[0.2,2x2x2,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
bspline-syn/per-stage | B-spline SyN | -i 1: the listing of the composite (a BSplineSmoothingOnUpdateDisplacementFieldTransform) for the next stagecommand lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -i 1 -t BSplineSyN[0.2,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t GaussianDisplacementField[0.2,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline-syn/restrict-deformation | B-spline SyN | -g 1x1x0command lineantsRegistration -d 3 -v 1 -o r -g 1x1x0 -t BSplineSyN[0.25,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/save-state | B-spline SyN | --save-state: BSplineSyN does not set the registration state (ANTS-50)command lineantsRegistration -d 3 -v 1 -o r -j state.h5 -t BSplineSyN[0.25,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both because the the output transform will contain the this processes initializer | 2/2 |
bspline-syn/initial-field | B-spline SyN | -r a displacement fieldcommand lineantsRegistration -d 3 -v 1 -o r -r warp-3d.nii.gz -t BSplineSyN[0.25,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
bspline-syn/initial-bspline | B-spline SyN | -r a B-spline transform: collapsed, ANTs loses the stage's fields (ANTS-76)command lineantsRegistration -d 3 -v 1 -o r -r bspline-3d.txt -t BSplineSyN[0.25,2x2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline-syn/icp | B-spline SyN | ICP alone: each fixed point's tangent-space derivative fitted by a B-splinecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t BSplineSyN[0.3,2x2x2,0,3] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/cc-and-icp | B-spline SyN | CC and ICP (ITK-33)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t BSplineSyN[0.3,2x2x2,0,3] -m 'CC[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,2]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,0.5]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/igdm | B-spline SyN | IGDM alonecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t BSplineSyN[0.3,2x2x2,0,3] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-syn/mesh-dimension | B-spline SyN | a 2D mesh in 3D: ERROR, the mesh sizes do not match the dimensioncommand lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
bspline-syn/order-0 | B-spline SyN | spline order 0: the B-spline filter throwscommand lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2x2x2,0,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
bspline-syn/bad-mesh | B-spline SyN | a mesh that does not convertcommand lineantsRegistration -d 3 -v 1 -o r -t BSplineSyN[0.25,2xa,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both ERROR: Unable to convert ‘a’ to type ‘f’ as float | 2/2 |
displacement-field/gaussian-cc | displacement-field stages | GaussianDisplacementField with CC (radius 2), 2 levels, warped imagescommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -t GaussianDisplacementField[0.25,3,0.5] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 8/8 |
displacement-field/gaussian-meansquares | displacement-field stages | MeanSquares, no total-field smoothingcommand lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,2,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-mattes | displacement-field stages | Mattes (24 bins)command lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-mi2 | displacement-field stages | MI2command lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,20]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
displacement-field/gaussian-gc | displacement-field stages | GC (ITK-30)command lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-demons | displacement-field stages | Demonscommand lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'Demons[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-multi | displacement-field stages | CC and Mattescommand lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,0.7,2]' -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.3,20]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-masks | displacement-field stages | -x [mask,mask]command lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-regular | displacement-field stages | regular sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,1,Regular,0.5]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-random | displacement-field stages | random sampling 0.6command lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24,Random,0.6]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-float | displacement-field stages | --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t GaussianDisplacementField[0.25,3,0.5] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-2d | displacement-field stages | 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t GaussianDisplacementField[0.3,2,0] -m 'CC[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-oblique | displacement-field stages | oblique gridscommand lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-affine-then | displacement-field stages | Affine then the field, collapsed: 0GenericAffine.mat, 1Warp (no inverse)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 8/8 |
displacement-field/gaussian-composite | displacement-field stages | -a 1: the composite holds the field; no inverse compositecommand lineantsRegistration -d 3 -v 1 -o r -a 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-no-collapse | displacement-field stages | -z 0command lineantsRegistration -d 3 -v 1 -o r -z 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
displacement-field/gaussian-restrict | displacement-field stages | -g 1x1x0command lineantsRegistration -d 3 -v 1 -o r -g 1x1x0 -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-icp | displacement-field stages | MeanSquares and ICP: each point's derivative at its virtual voxelcommand lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -m 'ICP[<generated: reg-mask>,<generated: reg-mask>,0.5]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-initial-field | displacement-field stages | -r a displacement fieldcommand lineantsRegistration -d 3 -v 1 -o r -r warp-3d.nii.gz -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/gaussian-print-similarity | displacement-field stages | --print-similarity-measure-interval: the observer's copy of the field fails (ANTS-75)command lineantsRegistration -d 3 -v 1 -o r --print-similarity-measure-interval 1 -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
displacement-field/gaussian-interval-volumes | displacement-field stages | --write-interval-volumes: the observer's copy of the field fails (ANTS-75)command lineantsRegistration -d 3 -v 1 -o r --write-interval-volumes 1 -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
displacement-field/gaussian-save-state | displacement-field stages | --save-state: no state (ANTS-50)command lineantsRegistration -d 3 -v 1 -o r -j state.h5 -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both because the the output transform will contain the this processes initializer | 2/2 |
displacement-field/gaussian-igdm | displacement-field stages | IGDM first: not implemented in ANTscommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t GaussianDisplacementField[0.25,3,0.5] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
displacement-field/gaussian-icp-first | displacement-field stages | ICP first: ANTs allocates the field on a null imagecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t GaussianDisplacementField[0.25,3,0.5] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both larmorx: antsRegistration: a displacement field stage whose first metric is a point-set metric: ANTs allocates the field on the null image of its first metric; ANTs crashes here, the replica stops | 4/4 |
displacement-field/gaussian-variance-missing | displacement-field stages | GaussianDisplacementField[0.25,3]: the total variance is requiredcommand lineantsRegistration -d 3 -v 1 -o r -t GaussianDisplacementField[0.25,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both ERROR: Unable to convert ‘’ to type ‘f’ as float | 2/2 |
displacement-field/bspline-meansquares | displacement-field stages | BSplineDisplacementField (mesh 3x3x3, doubled at the first level), MeanSquares, warpedcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -t BSplineDisplacementField[0.25,3x3x3,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 8/8 |
displacement-field/bspline-cc | displacement-field stages | CCcommand lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.25,3x3x3,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-total-mesh | displacement-field stages | a total-field mesh (4x4x4, update 5x5x5)command lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.25,5x5x5,4x4x4,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-coarse-total | displacement-field stages | a total mesh of 1: the stationary boundary dominates the fit, the scales fall below epsilon and the stage failscommand lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.3,3x3x2,1x1x1,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
displacement-field/bspline-order-2 | displacement-field stages | order 2command lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.25,3x3x3,0,2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-knot | displacement-field stages | a knot spacing (12 mm)command lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.3,12,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-mattes | displacement-field stages | Mattescommand lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.25,3x3x3,0,3] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-masks | displacement-field stages | -x [mask,mask]command lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t BSplineDisplacementField[0.25,3x3x3,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-float | displacement-field stages | --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t BSplineDisplacementField[0.25,3x3x3,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-2d | displacement-field stages | 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t BSplineDisplacementField[0.3,3x3,0,3] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
displacement-field/bspline-mesh-dimension | displacement-field stages | a 2D mesh in 3D: ERRORcommand lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.25,3x3,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
displacement-field/bspline-order-0 | displacement-field stages | order 0: the B-spline filter throwscommand lineantsRegistration -d 3 -v 1 -o r -t BSplineDisplacementField[0.25,3x3x3,0,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
displacement-field/gaussian-then-bspline | displacement-field stages | the two field stages, collapsed into one fieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t GaussianDisplacementField[0.25,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox -t BSplineDisplacementField[0.25,3x3x3,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [4x3,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline/meansquares | B-spline stages | BSpline (mesh 2x2x2, doubled at each level), MeanSquares, 2 levels, warpedcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 8/8 |
bspline/cc | B-spline stages | CC (radius 2)command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/mattes | B-spline stages | Mattes (24 bins)command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/mi2 | B-spline stages | MI2command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,20]' -c [3,1e-6,3] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in r0BSpline.txt; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
bspline/gc | B-spline stages | GCcommand lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/demons | B-spline stages | Demons: the metric needs a displacement field (ITK's exception)command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'Demons[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
bspline/multi | B-spline stages | MeanSquares and CCcommand lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,0.5,2]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/masks | B-spline stages | -x [mask,mask]command lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/regular | B-spline stages | regular sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,1,Regular,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/random | B-spline stages | random sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,1,Random,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/float | B-spline stages | --float 1, 2 levelscommand lineantsRegistration -d 3 -v 1 -o r --float 1 -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/2d | B-spline stages | 2Dcommand lineantsRegistration -d 2 -v 1 -o r -t BSpline[0.2,3x2] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/oblique | B-spline stages | oblique gridscommand lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/knot-spacing | B-spline stages | a knot spacing (30 mm)command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,30] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/mesh-1 | B-spline stages | mesh 1x1x1command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,1x1x1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/icp | B-spline stages | MeanSquares and ICPcommand lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -m 'ICP[<generated: reg-mask>,<generated: reg-mask>,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/affine-then | B-spline stages | Affine then BSpline, collapsed: 0GenericAffine.mat, 1BSpline.txtcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 8/8 |
bspline/composite | B-spline stages | -a 1: the composite holds the B-spline transform; no inversecommand lineantsRegistration -d 3 -v 1 -o r -a 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/no-collapse | B-spline stages | -z 0command lineantsRegistration -d 3 -v 1 -o r -z 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline/per-stage | B-spline stages | -i 1 -a 1 -z 0: ANTs cannot initialise a B-spline transform from the affine onecommand lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -i 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/then-affine | B-spline stages | BSpline then Affine: no Jacobian with respect to position; the Affine stage fails and is dropped (ANTS-49)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline/then-bspline | B-spline stages | two BSpline stages: the second fails (no Jacobian with respect to position)command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t BSpline[0.2,1x1x1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
bspline/print-similarity | B-spline stages | --print-similarity-measure-interval: the observer's copy of the transform fails (ANTS-75)command lineantsRegistration -d 3 -v 1 -o r --print-similarity-measure-interval 1 -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
bspline/print-similarity-mesh-1 | B-spline stages | --print-similarity-measure-interval with mesh 1: the copy workscommand lineantsRegistration -d 3 -v 1 -o r --print-similarity-measure-interval 1 -t BSpline[0.2,1x1x1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/restrict | B-spline stages | -g 1x1x0: weights for the dimension, not the parameters (ITK's exception)command lineantsRegistration -d 3 -v 1 -o r -g 1x1x0 -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
bspline/save-state | B-spline stages | --save-state: no state (ANTS-50)command lineantsRegistration -d 3 -v 1 -o r -j state.h5 -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both because the the output transform will contain the this processes initializer | 2/2 |
bspline/initial-bspline-syn | B-spline stages | -r a B-spline transform, then SyN: collapsed, ANTs loses the SyN fields (ANTS-76)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -r bspline-3d.txt -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline/initial-bspline-syn-composite | B-spline stages | -r an affine and B-spline composite, then SyN, -a 1command lineantsRegistration -d 3 -v 1 -o r -a 1 -r bspline-3d.h5 -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
bspline/initial-bspline-affine | B-spline stages | -r a B-spline transform, then Affine: the stage fails and is dropped (ANTS-49)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -r bspline-3d.txt -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline/fixed-bspline | B-spline stages | -q a B-spline transformcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -q bspline-3d.txt -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline/initial-bspline-inverse | B-spline stages | -r [bspline,1]: a B-spline transform has no inversecommand lineantsRegistration -d 3 -v 1 -o r -r [bspline-3d.txt,1] -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
bspline/restore-bspline | B-spline stages | --restore-state a B-spline transform, then SyN (ANTS-76)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -k bspline-3d.txt -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline/interval-volumes-mesh-1 | B-spline stages | --write-interval-volumes with mesh 1: the copies work, the volumes through the B-spline transformcommand lineantsRegistration -d 3 -v 1 -o r --write-interval-volumes 1 -t BSpline[0.2,1x1x1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
bspline/mesh-0 | B-spline stages | a mesh size of 0: ANTs divides by zero and fails on garbage indices (ANTS-77)command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,0x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Expected divergence ANTs computes a grid of NaN and infinite geometry, prints vnl’s SVD warnings and fails with an exception on garbage indices (ANTS-77); the replica stops with an error after the same messages | 1/1 |
bspline/mesh-short | B-spline stages | a 2D mesh in 3D: ANTs reads past the end of the list (ANTS-77)command lineantsRegistration -d 3 -v 1 -o r -t BSpline[0.2,2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Expected divergence ANTs reads the missing mesh size past the end of its vector (here a 0) and fails as with a mesh size of 0 (ANTS-77); the replica stops with an error | 1/1 |
bspline/icp-first | B-spline stages | ICP first: ANTs reads the null image of the first metriccommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t BSpline[0.2,2x2x2] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both larmorx: antsRegistration: a BSpline stage whose first metric is a point-set metric: ANTs reads the spacing and size of the null image of its first metric; ANTs crashes here, the replica stops | 4/4 |
exponential/meansquares | exponential stages | Exponential (update variance 3, velocity variance 0.5), MeanSquares, 2 levels, warped imagescommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 10/10 |
exponential/no-collapse | exponential stages | -z 0 after an affine stage: 0Affine.mat, 1Warp, 1InverseWarp and 1VelocityFieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 11/11 |
exponential/short-name | exponential stages | -t Exp[...] with -z 0: no InverseWarp file (ANTS-81)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -t Exp[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/collapse | exponential stages | an affine stage, then the velocity field: collapsedcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
exponential/composite | exponential stages | -a 1command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -a 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/integration-steps | exponential stages | 4 integration steps, no velocity smoothingcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,2,0,4] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/sigmas-in-mm | exponential stages | smoothing sigmas given in mm (as ANTs takes voxels, ANTS-80)command lineantsRegistration -d 3 -v 1 -o r -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 2x0mm | Bit-identical exit status, standard output, standard error, every file | 7/7 |
exponential/cc | exponential stages | CC (radius 2)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/mattes | exponential stages | Mattes (24 bins)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
exponential/mi2 | exponential stages | MI2command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'MI2[<generated: reg-fixed>,<generated: reg-moving>,1,20]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
exponential/gc | exponential stages | GC (ITK-30)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'GC[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/demons | exponential stages | Demonscommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'Demons[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/multi | exponential stages | CC (0.7) and Mattes (0.3)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,0.7,2]' -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.3,20]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/masks | exponential stages | -x [mask,mask]command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -x '[<generated: reg-mask>,<generated: reg-mask>]' -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/float | exponential stages | --float 1command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] --float 1 -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/2d | exponential stages | 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/oblique | exponential stages | oblique gridscommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/restrict | exponential stages | -g 1x1x0command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -g 1x1x0 -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/initial-field | exponential stages | -r a displacement fieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -r warp-3d.nii.gz -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
exponential/per-stage | exponential stages | -i 1 -a 1 -z 0 after an affine stagecommand lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -i 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
exponential/then-affine | exponential stages | an affine stage after the velocity fieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox | Expected divergence the affine stage after a displacement field samples its scales’ 1000 voxels from the clock (ANTS-72): two ANTs runs differ | 1/1 |
exponential/regular-sampling | exponential stages | MeanSquares, regular sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,1,Regular,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Expected divergence the stage never sets its metric’s sampling seed (ANTS-82): ANTs samples from the clock, two runs differ; the replica samples with the registration’s seed | 1/1 |
exponential/icp-second | exponential stages | MeanSquares and ICPcommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
exponential/print-similarity | exponential stages | --print-similarity-measure-interval: the observer's copy of the field fails (ANTS-75)command lineantsRegistration -d 3 -v 1 -o r --print-similarity-measure-interval 1 -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
exponential/interval-volumes | exponential stages | --write-interval-volumes: the observer's copy of the field fails (ANTS-75)command lineantsRegistration -d 3 -v 1 -o r --write-interval-volumes 1 -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
exponential/save-state | exponential stages | --save-state: no state (ANTS-50)command lineantsRegistration -d 3 -v 1 -o r -j state.h5 -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both because the the output transform will contain the this processes initializer | 2/2 |
exponential/zero-iterations | exponential stages | no iteration and one level: the transform never integrates, its inverse has no field (ITK-41)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [0,1e-6,3] -f 1 -s 0 | Rejected by both Description: ITK ERROR: ConstantVelocityFieldTransform(0x6223b4403680): No displacement field is specified. | 2/2 |
exponential/zero-iterations-prefix | exponential stages | no iteration, no warped images: written as ANTs writes it (ITK-41)command lineantsRegistration -d 3 -v 1 -o r -t Exponential[0.5,3,0.5] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [0,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
exponential/icp-first | exponential stages | ICP first: ANTs allocates the velocity field on a null imagecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t Exponential[0.5,3,0.5] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both larmorx: antsRegistration: an Exponential stage whose first metric is a point-set metric: ANTs allocates the velocity field on the null image of its first metric; ANTs crashes here, the replica stops | 4/4 |
exponential/missing-parameter | exponential stages | Exponential[0.5,3]: the velocity variance is requiredcommand lineantsRegistration -d 3 -v 1 -o r -t Exponential[0.5,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both ERROR: Unable to convert ‘’ to type ‘f’ as float | 2/2 |
bspline-exponential/meansquares | B-spline exponential stages | BSplineExponential (update mesh 4x4x4), MeanSquares, 2 levels, warped imagescommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t BSplineExponential[0.5,4x4x4,0,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 10/10 |
bspline-exponential/velocity-mesh | B-spline exponential stages | a velocity-field mesh (2x2x2) and 5 integration stepscommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t BSplineExponential[0.5,4x4x4,2x2x2,5,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
bspline-exponential/knot-spacing | B-spline exponential stages | a knot spacing (30 mm), -z 0command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -t BSplineExponential[0.5,30,0,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
bspline-exponential/cc | B-spline exponential stages | CCcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t BSplineExponential[0.5,4x4x4,0,0,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
bspline-exponential/float | B-spline exponential stages | --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t BSplineExponential[0.5,4x4x4,0,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-exponential/2d | B-spline exponential stages | 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t BSplineExponential[0.5,4x4,2x2,0,3] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
bspline-exponential/masks | B-spline exponential stages | -x [mask,mask]command lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t BSplineExponential[0.5,4x4x4,0,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-exponential/order-2 | B-spline exponential stages | order 2command lineantsRegistration -d 3 -v 1 -o r -t BSplineExponential[0.5,4x4x4,0,0,2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
bspline-exponential/mesh-dimension | B-spline exponential stages | a 2D mesh in 3D: ERRORcommand lineantsRegistration -d 3 -v 1 -o r -t BSplineExponential[0.5,4x4,0,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
bspline-exponential/random-sampling | B-spline exponential stages | Mattes, random sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t BSplineExponential[0.5,4x4x4,0,0,3] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24,Random,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Expected divergence the stage never sets its metric’s sampling seed (ANTS-82): ANTs samples from the clock, two runs differ; the replica samples with the registration’s seed | 1/1 |
time-varying/meansquares | time-varying velocity fields | TimeVaryingVelocityField (3 time indices), MeanSquares, 2 levels, -z 0: Warp, InverseWarp, VelocityFieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 11/11 |
time-varying/short-name | time-varying velocity fields | tvf[...] with time variances (0.5, 0.25)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -t tvf[0.5,3,3,0.5,0.5,0.25] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 10/10 |
time-varying/collapse | time-varying velocity fields | an affine stage, then the velocity field: collapsedcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 10/10 |
time-varying/composite | time-varying velocity fields | -a 1 after an affine stagecommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -a 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/cc | time-varying velocity fields | CCcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/mattes | time-varying velocity fields | Mattes (24 bins)command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,1,24]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying/multi | time-varying velocity fields | CC (0.7) and Mattes (0.3)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingVelocityField[0.5,3,3,0,0,0] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,0.7,2]' -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.3,20]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/masks | time-varying velocity fields | -x [mask,mask]: the masks given to the metric unchanged (ITK-40)command lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying/float | time-varying velocity fields | --float 1command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] --float 1 -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/2d | time-varying velocity fields | 2D, 4 time indicescommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -z 0 -t TimeVaryingVelocityField[0.5,4,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -c [3x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 10/10 |
time-varying/one-time-index | time-varying velocity fields | one time indexcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingVelocityField[0.5,1,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/restrict | time-varying velocity fields | -g 1x0x1command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -g 1x0x1 -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/two-stages | time-varying velocity fields | two TimeVaryingVelocityField stages: the number of time indices from the last -t (ANTS-79)command lineantsRegistration -d 3 -v 1 -o r -z 0 -t TimeVaryingVelocityField[0.5,2,3,0,0,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t TimeVaryingVelocityField[0.5,3,3,0,0,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 11/11 |
time-varying/then-syn | time-varying velocity fields | a SyN stage after the velocity fieldcommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingVelocityField[0.5,2,3,0,0,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t SyN[0.25,3,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/per-stage | time-varying velocity fields | -i 1 -a 1 -z 0 after an affine stagecommand lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -i 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying/fixed-initial | time-varying velocity fields | -q an affine transform: ignored by the method (ITK-39)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -q initial.mat -t TimeVaryingVelocityField[0.5,2,3,0,0,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying/print-similarity | time-varying velocity fields | --print-similarity-measure-interval: no full-scale similarity in this observercommand lineantsRegistration -d 3 -v 1 -o r --print-similarity-measure-interval 1 -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying/icp-second | time-varying velocity fields | CC and ICPcommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t TimeVaryingVelocityField[0.5,2,3,0,0,0] -m 'CC[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,2]' -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
time-varying/regular-sampling | time-varying velocity fields | MeanSquares, regular sampling 0.5command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1,1,Regular,0.5]' -c [3,1e-6,3] -f 1 -s 0 | Expected divergence the stage never sets its metric’s sampling seed (ANTS-82): ANTs samples from the clock, two runs differ; the replica samples with the registration’s seed | 1/1 |
time-varying/zero-iterations-prefix | time-varying velocity fields | no iteration: the transform never integrates; no warped images (ITK-41)command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [0,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying/zero-iterations | time-varying velocity fields | no iteration and warped images: ANTs maps points through a field without an interpolator input (ITK-41)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [0,1e-6,3] -f 1 -s 0 | Rejected by both larmorx: upstream defect ITK-41 (ITK) reproduced: stage 0: the time-varying velocity field transform never integrated its velocity field: ITK crashes at the next use of the transform [LARMORX_UPSTREAM | 4/4 |
time-varying/zero-time-indices | time-varying velocity fields | 0 time indices: ANTs divides by 0command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingVelocityField[0.5,0,3,0,0,0] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both larmorx: antsRegistration: a TimeVaryingVelocityField stage with 0 time indices: ANTs divides the velocity field’s voxels by 0; ANTs crashes here, the replica stops | 4/4 |
time-varying/icp-first | time-varying velocity fields | ICP first: ANTs allocates the velocity field on a null imagecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t TimeVaryingVelocityField[0.5,3,3,0,0.5,0] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both larmorx: antsRegistration: a TimeVaryingVelocityField stage whose first metric is a point-set metric: ANTs allocates the velocity field on the null image of its first metric; ANTs crashes here, larmor | 4/4 |
time-varying/missing-parameter | time-varying velocity fields | TimeVaryingVelocityField[0.5,2,3]: the other variances are requiredcommand lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingVelocityField[0.5,2,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 | Rejected by both ERROR: Unable to convert ‘’ to type ‘f’ as float | 2/2 |
time-varying-bspline/meansquares | time-varying B-spline velocity fields | TimeVaryingBSplineVelocityField (mesh 4x4x4x2, 4 time samples), MeanSquares, 2 levels, warped imagescommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 10/10 |
time-varying-bspline/float | time-varying B-spline velocity fields | --float 1command lineantsRegistration -d 3 -v 1 -o r --float 1 -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/2d | time-varying B-spline velocity fields | 2Dcommand lineantsRegistration -d 2 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingBSplineVelocityField[0.5,4x4x2,4,3] -m 'MeanSquares[<generated: reg-fixed-2d>,<generated: reg-moving-2d>,1]' -c [2x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying-bspline/masks | time-varying B-spline velocity fields | -x [mask,mask]command lineantsRegistration -d 3 -v 1 -o r -x '[<generated: reg-mask>,<generated: reg-mask>]' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/fixed-mask-oblique | time-varying B-spline velocity fields | a fixed mask on an oblique gridcommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: reg-mask-oblique>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/cc | time-varying B-spline velocity fields | CCcommand lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,1,2]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/multi | time-varying B-spline velocity fields | CC (0.7) and Mattes (0.3)command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'CC[<generated: reg-fixed>,<generated: reg-moving>,0.7,2]' -m 'Mattes[<generated: reg-fixed>,<generated: reg-moving>,0.3,20]' -c [2,1e-6,3] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in r0InverseWarp.nii.gz, r0Warp.nii.gz; identical to ANTs with CORE-MATH preloaded (every output) | 13/13 |
time-varying-bspline/order-2 | time-varying B-spline velocity fields | order 2command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/samples-2 | time-varying B-spline velocity fields | 2 time samples, mesh 3x3x3x1command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,3x3x3x1,2,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/samples-1 | time-varying B-spline velocity fields | 1 time samplecommand lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,3x3x3x1,1,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/oblique | time-varying B-spline velocity fields | oblique gridscommand lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'MeanSquares[<generated: reg-fixed-oblique>,<generated: reg-moving-oblique>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/restrict | time-varying B-spline velocity fields | -g 1x1x0command lineantsRegistration -d 3 -v 1 -o r -g 1x1x0 -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/interval-volumes | time-varying B-spline velocity fields | --write-interval-volumes: not written by this observercommand lineantsRegistration -d 3 -v 1 -o r --write-interval-volumes 1 -t TimeVaryingBSplineVelocityField[0.5,3x3x3x1,2,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/per-stage | time-varying B-spline velocity fields | -i 1 -a 1 -z 0 after an affine stagecommand lineantsRegistration -d 3 -v 1 -o r -a 1 -z 0 -i 1 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [6,1e-6,5] -f 1 -s 0vox -t TimeVaryingBSplineVelocityField[0.5,3x3x3x1,2,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/zero-iterations | time-varying B-spline velocity fields | no iteration on two levelscommand lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz,iw.nii.gz] -t TimeVaryingBSplineVelocityField[0.5,3x3x3x1,2,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [0x1,1e-6,3] -f 2x1 -s 0x0 | Bit-identical exit status, standard output, standard error, every file | 9/9 |
time-varying-bspline/icp | time-varying B-spline velocity fields | ICP alone (identity-direction phantoms): the points fitted directlycommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask-lpsid>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'ICP[<generated: ps-labels-fixed-lpsid>,<generated: ps-labels-moving-lpsid>,1]' -c [2x2,1e-6,3] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/pse | time-varying B-spline velocity fields | PSE alonecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask-lpsid>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'PSE[<generated: ps-labels-fixed-lpsid>,<generated: ps-labels-moving-lpsid>,1,1,0,2,5]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/jhct | time-varying B-spline velocity fields | JHCT alonecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask-lpsid>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'JHCT[<generated: ps-labels-fixed-lpsid>,<generated: ps-labels-moving-lpsid>,1,1,0,2,5,1.1,0]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/igdm | time-varying B-spline velocity fields | IGDM alonecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask-lpsid>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'IGDM[<generated: ps-intensity-fixed-lpsid>,<generated: ps-intensity-moving-lpsid>,1,<generated: ps-mask-lpsid>,<generated: ps-mask-moving-lpsid>]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/cc-and-icp | time-varying B-spline velocity fields | CC and ICP: the image metric warped twice (ITK-36)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask-lpsid>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'CC[<generated: ps-intensity-fixed-lpsid>,<generated: ps-intensity-moving-lpsid>,1,2]' -m 'ICP[<generated: ps-labels-fixed-lpsid>,<generated: ps-labels-moving-lpsid>,0.5]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/meansquares-icp-cc | time-varying B-spline velocity fields | MeanSquares, ICP and CC (ITK-36, ITK-37)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask-lpsid>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'MeanSquares[<generated: ps-intensity-fixed-lpsid>,<generated: ps-intensity-moving-lpsid>,1]' -m 'ICP[<generated: ps-labels-fixed-lpsid>,<generated: ps-labels-moving-lpsid>,0.5]' -m 'CC[<generated: ps-intensity-fixed-lpsid>,<generated: ps-intensity-moving-lpsid>,1,2]' -c [2,1e-6,3] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 7/7 |
time-varying-bspline/icp-ras | time-varying B-spline velocity fields | ICP alone on RAS-stored phantoms: ANTs fits the points outside the parametric domain and throws (ITK-38)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'ICP[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1]' -c [2x2,1e-6,3] -f 2x1 -s 1x0vox | Rejected by both 1 | 2/2 |
time-varying-bspline/icp-then-cc | time-varying B-spline velocity fields | ICP first, then CC: ANTs cannot convert the multi-metric queuecommand lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask-lpsid>' -t TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] -m 'ICP[<generated: ps-labels-fixed-lpsid>,<generated: ps-labels-moving-lpsid>,0.5]' -m 'CC[<generated: ps-intensity-fixed-lpsid>,<generated: ps-intensity-moving-lpsid>,1,2]' -c [2,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
time-varying-bspline/mesh-dimension | time-varying B-spline velocity fields | a mesh without its time size: the message, then failurecommand lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,4x4x4,4,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Rejected by both 1 | 2/2 |
time-varying-bspline/mesh-0 | time-varying B-spline velocity fields | a mesh size of 0: ANTs divides by zero and fails on a parametric domain of no span (ANTS-77)command lineantsRegistration -d 3 -v 1 -o r -t TimeVaryingBSplineVelocityField[0.5,3x0x3x1,2,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [2,1e-6,3] -f 1 -s 0 | Expected divergence ANTs computes control-point spacings of infinity, prints vnl’s SVD warnings and fails with an exception on a parametric domain of no span (ANTS-77); the replica stops with an error after the same messages | 1/1 |
corrections/igdm-k-neighbourhood | corrections of defects | IGDM with kNeighborhood 5 and gradientSigma 1.5, given sigmas (ANTS-67, ANTS-68)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>,0x0x0,50,5,5,1.5]' -c [20,1e-8,10] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
corrections/igdm-true-gradient | corrections of defects | IGDM, given sigmas, one neighbour, central differences: the derivative alone (ANTS-68)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-moving>,1,<generated: ps-mask>,<generated: ps-mask-moving>,0x0x0,50,5,1,0]' -c [20,1e-8,10] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
corrections/igdm-same-sets | corrections of defects | IGDM of an image with itself moved by half a voxel: every nearest distance 1 mm, the distance sigma 0, NaN parameters, the stage dropped (ANTS-70)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -r half-voxel-x.txt -t Rigid[0.1] -m 'IGDM[<generated: ps-intensity-fixed>,<generated: ps-intensity-fixed>,1,<generated: ps-mask>,<generated: ps-mask>]' -c [10,1e-8,10] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
corrections/jhct-isotropic-sigma-2 | corrections of defects | JHCT, isotropic, sigma 2: the total number of points alone (ITK-24)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,2,50,1.1,0]' -c [20,1e-8,10] -f 1 -s 0 | Identical with correctly rounded maths differs from ANTs as built in r0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
corrections/jhct-anisotropic | corrections of defects | JHCT, anisotropic: the Parzen kernel sigma (ITK-29, ITK-24, ITK-26)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: ps-mask>' -t Rigid[0.1] -m 'JHCT[<generated: ps-labels-fixed>,<generated: ps-labels-moving>,1,1,0,1,50,1.1,1]' -c [20,1e-8,10] -f 1 -s 0 | Bit-identical exit status, standard output, standard error, every file | 6/6 |
corrections/bspline-then-affine | corrections of defects | an affine stage after a BSpline stage, with a mask (ITK-35; ANTs drops it, ANTS-49)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -x '<generated: reg-mask>' -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t Affine[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [20,1e-8,10] -f 1 -s 0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
corrections/bspline-then-bspline | corrections of defects | two BSpline stages (ITK-35; ANTs fails)command lineantsRegistration -d 3 -v 1 -o r -x '<generated: reg-mask>' -t BSpline[0.2,2x2x2] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [3,1e-6,3] -f 1 -s 0 -t BSpline[0.2,1x1x1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [20,1e-6,5] -f 1 -s 0 | Rejected by both 1 | 2/2 |
corrections/initial-bspline-then-rigid | corrections of defects | -r a B-spline transform, then Rigid, with a mask (ITK-35)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -x '<generated: reg-mask>' -r bspline-3d.txt -t Rigid[0.1] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [20,1e-8,10] -f 1 -s 0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
corrections/bspline-field-coarse | corrections of defects | BSplineDisplacementField 3x3x3 over 2 levels (ANTS-78; corrected, the coarse level needs ITK-42)command lineantsRegistration -d 3 -v 1 -o [r,w.nii.gz] -t BSplineDisplacementField[0.3,3x3x3,0,3] -m 'MeanSquares[<generated: reg-fixed>,<generated: reg-moving>,1]' -c [5x3,1e-6,5] -f 2x1 -s 1x0vox | Bit-identical exit status, standard output, standard error, every file | 7/7 |
real/ds000005-sub-01/boldref-affine | real images: fMRIPrep's calls | fMRIPrep 25.2.6's enhance_and_skullstrip_bold_wf/norm of ds000005-sub-01-noreconallcommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: norm>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
real/ds005040-sub-002/boldref-affine | real images: fMRIPrep's calls | fMRIPrep 25.2.6's enhance_and_skullstrip_bold_wf/norm of ds005040-sub-002-noreconallcommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005040-sub-002-noreconall: norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
real/ds005040-sub-002/coregister | real images: fMRIPrep's calls | fMRIPrep 25.2.6's fmapreg_wf/coregister of ds005040-sub-002-noreconallcommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005040-sub-002-noreconall: coregister>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
real/ds005454-sub-16/coregister | real images: fMRIPrep's calls | fMRIPrep 25.2.6's fmapreg_wf/coregister of ds005454-sub-16-noreconallcommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005454-sub-16-noreconall: coregister>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
real-syn-small/ds000005-sub-01/brain-extraction | real images: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's norm of ds000005-sub-01-noreconall, whole (Rigid, Affine, SyN), on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 1.63 in a parameter | 18/18 |
real-linear/ds000005-sub-01/template | real images: the template registration's linear stages | the linear stages of fMRIPrep 25.2.6's registration of ds000005-sub-01-noreconall (its SyN stage left out)command lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: registration>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
real-syn-small/ds000005-sub-01/template worked example | real images: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's registration of ds000005-sub-01-noreconall, whole (Rigid, Affine, SyN), on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: registration>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 0.945 in a parameter | 16/16 |
real-bsplinesyn-small/ds000005-sub-01/template worked example | real images: B-spline SyN, every third voxel | fMRIPrep 25.2.6's registration of ds000005-sub-01-noreconall with BSplineSyN[0.1,26,0,3] (antsRegistrationSyN.sh's) for its SyN stage, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: registration>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 0.51 in a parameter | 16/16 |
real-exponential-small/ds000005-sub-01/template | real images: an exponential stage, every third voxel | fMRIPrep 25.2.6's registration of ds000005-sub-01-noreconall with Exponential[0.25,3,0.5] for its SyN stage, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: registration>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 0.113 in a parameter | 16/16 |
real-bspline-exponential-small/ds000005-sub-01/template | real images: a B-spline exponential stage, every third voxel | fMRIPrep 25.2.6's registration of ds000005-sub-01-noreconall with BSplineExponential[0.25,26,0,0,3] for its SyN stage, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: registration>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
real-time-varying-small/ds000005-sub-01/template worked example | real images: a time-varying velocity field, every third voxel | fMRIPrep 25.2.6's registration of ds000005-sub-01-noreconall with TimeVaryingVelocityField[0.5,4,3,0,0,0] for its SyN stage, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: registration>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 0.265 in a parameter | 16/16 |
real-time-varying-bspline-small/ds000005-sub-01/template | real images: a time-varying B-spline velocity field, every third voxel | fMRIPrep 25.2.6's registration of ds000005-sub-01-noreconall with TimeVaryingBSplineVelocityField[0.5,4x4x4x2,4,3] for its SyN stage, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: registration>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
difftest/empty-sample-set-exception | random testing | random testing's permanent case: a stage whose sampled point set is empty: ITK's exception at itkImageToImageMetricv4.hxx:263 with the metric's class (the replica printed line 268 and the base class) | Bit-identical exit status, standard output, standard error, every file | 6/6 |
difftest/field-under-a-mat-name | random testing | random testing's permanent case: -z 0 with a Translation stage whose Demons metric throws (no transform added) before a Gaussian displacement-field stage: ANTs writes the field under the Translation stage's .mat name as its own GaussianSmoothingOnUpdateDisplacementFieldTransform; the replica wrote a plain DisplacementFieldTransform | Bit-identical exit status, standard output, standard error, every file | 7/7 |
difftest/float-field-jacobian-index | random testing | random testing's permanent case: an Exponential stage after a deformable one with --float 1: the displacement field's Jacobian at the voxel nearest a float point, indexed in float (the replica indexed in double; half-way points rounded differently) | Bit-identical exit status, standard output, standard error, every file | 7/7 |
difftest/four-dimensional-field-parameters | random testing | random testing's permanent case: a 4D B-spline displacement-field stage: the runner reads the 4D composite by its HDF5 parameters | Bit-identical exit status, standard output, standard error, every file | 7/7 |
difftest/mattes-later-level-one-intensity | random testing | random testing's permanent case: a BSpline stage with MI and a sparse mask where the full-resolution fixed image has one intensity: ITK keeps the first level's joint histogram images, so the second level's bin size of 0 is used silently (the metric 0) where the first level would refuse (ITK-100, reproduced) | Bit-identical exit status, standard output, standard error, every file | 7/7 |
difftest/mattes-nan-bin-svd-warning | random testing | random testing's permanent case: Mattes on images with NaN values (a time-varying velocity field stage): the joint histogram's NaN bin size makes vnl's SVD fail and print its warning, once per stage (the replica printed none). Dense sampling (`None`) since 2026-10-28: this stage seeds `Random` and `Regular` sampling from the clock in ANTs (ANTS-82), and its metric values then vary from run to run (0 or ±2.2e-16 here); found with `Random,0.25` | Bit-identical exit status, standard output, standard error, every file | 7/7 |
difftest/sampled-derivatives-aliased | random testing | random testing's permanent case: a sampled B-spline displacement-field stage with a shrink factor: the derivatives of sample points outside the shrunk virtual region, aliased by ITK (ITK-97, reproduced) | Bit-identical exit status, standard output, standard error, every file | 7/7 |
difftest/syn-float-metric-sum-overflow | random testing | random testing's permanent case: B-spline SyN with --float 1 and a mask leaving no valid point: ITK sums the two halves' FLT_MAX metric values in float (inf) before halving them; the replica summed in double | Bit-identical exit status, standard output, standard error, every file | 9/9 |
panel/fmriprep/ds000005-sub-01/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds000005-sub-01 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds000005-sub-01/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds000005-sub-01 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 3.18 in a parameter | 18/18 |
panel/fmriprep/ds000005-sub-01/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds000005-sub-01 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output); the replica − ANTs at 1 thread: 0.614 in a parameter | 16/16 |
panel/fmriprep/ds000005-sub-01/bold[task_mixedgamblestask_run_01]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_mixedgamblestask_run_01]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds000005-sub-01 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
panel/fmriprep/ds000005-sub-01/bold[task_mixedgamblestask_run_02]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_mixedgamblestask_run_02]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds000005-sub-01 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds000005-sub-01/bold[task_mixedgamblestask_run_03]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_mixedgamblestask_run_03]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds000005-sub-01 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000005-sub-01-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds000228-sub-pixar008/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds000228-sub-pixar008 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000228-sub-pixar008-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds000228-sub-pixar008/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds000228-sub-pixar008 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000228-sub-pixar008-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds000228-sub-pixar008/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds000228-sub-pixar008 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000228-sub-pixar008-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds000228-sub-pixar008/bold[task_pixar]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_pixar]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds000228-sub-pixar008 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000228-sub-pixar008-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds000256-sub-CTS203/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds000256-sub-CTS203 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000256-sub-CTS203-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds000256-sub-CTS203/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds000256-sub-CTS203 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000256-sub-CTS203-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds000256-sub-CTS203/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds000256-sub-CTS203 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000256-sub-CTS203-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds000256-sub-CTS203/bold[task_restbaseline]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_restbaseline]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds000256-sub-CTS203 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000256-sub-CTS203-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds000256-sub-CTS203/bold[task_restfixationfixed]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_restfixationfixed]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds000256-sub-CTS203 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000256-sub-CTS203-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds000256-sub-CTS203/bold[task_restfixationnofeedback]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_restfixationnofeedback]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds000256-sub-CTS203 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds000256-sub-CTS203-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds001600-sub-1/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds001600-sub-1 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds001600-sub-1-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds001600-sub-1/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds001600-sub-1 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds001600-sub-1-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds001600-sub-1/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds001600-sub-1 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds001600-sub-1-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds002422-sub-02/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds002422-sub-02 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds002422-sub-02/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds002422-sub-02 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds002422-sub-02/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds002422-sub-02 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds002422-sub-02/bold[task_arithm_run_01]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_arithm_run_01]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds002422-sub-02 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds002422-sub-02/bold[task_arithm_run_01]/bold_fit/fmapreg/coregister | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_arithm_run_01]/bold_fit/fmapreg/coregister` of the fMRIPrep run ds002422-sub-02 (bold_wf/bold_fit_wf/fmapreg_wf/coregister): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: fmapreg_wf/coregister>' | Bit-identical exit status, standard output, standard error, every file; the replica − ANTs at 1 thread: 0 in a parameter | 7/7 |
panel/fmriprep/ds002422-sub-02/bold[task_arithm_run_02]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_arithm_run_02]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds002422-sub-02 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds002422-sub-02/bold[task_arithm_run_02]/bold_fit/fmapreg/coregister | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_arithm_run_02]/bold_fit/fmapreg/coregister` of the fMRIPrep run ds002422-sub-02 (bold_wf/bold_fit_wf/fmapreg_wf/coregister): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: fmapreg_wf/coregister>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds002422-sub-02/bold[task_rest]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_rest]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds002422-sub-02 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds002422-sub-02/bold[task_rest]/bold_fit/fmapreg/coregister | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_rest]/bold_fit/fmapreg/coregister` of the fMRIPrep run ds002422-sub-02 (bold_wf/bold_fit_wf/fmapreg_wf/coregister): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002422-sub-02-noreconall: fmapreg_wf/coregister>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds002603-sub-01/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds002603-sub-01 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002603-sub-01-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds002603-sub-01/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds002603-sub-01 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002603-sub-01-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds002603-sub-01/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds002603-sub-01 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002603-sub-01-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds002603-sub-01/bold[task_wm_acq_Mb4Mm3Tr550]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_wm_acq_Mb4Mm3Tr550]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds002603-sub-01 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002603-sub-01-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds002603-sub-01/bold[task_wm_acq_Mb4Mm3Tr550]/bold_fit/fmapreg/coregister | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_wm_acq_Mb4Mm3Tr550]/bold_fit/fmapreg/coregister` of the fMRIPrep run ds002603-sub-01 (bold_wf/bold_fit_wf/fmapreg_wf/coregister): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002603-sub-01-noreconall: fmapreg_wf/coregister>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds002603-sub-01/bold[task_wm_acq_SeqMm3Tr2000]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_wm_acq_SeqMm3Tr2000]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds002603-sub-01 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002603-sub-01-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds002603-sub-01/bold[task_wm_acq_SeqMm3Tr2000]/bold_fit/fmapreg/coregister | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_wm_acq_SeqMm3Tr2000]/bold_fit/fmapreg/coregister` of the fMRIPrep run ds002603-sub-01 (bold_wf/bold_fit_wf/fmapreg_wf/coregister): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds002603-sub-01-noreconall: fmapreg_wf/coregister>' | Identical with correctly rounded maths differs from ANTs as built in transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds003592-sub-280/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds003592-sub-280 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds003592-sub-280-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds003592-sub-280/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds003592-sub-280 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds003592-sub-280-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds003592-sub-280/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds003592-sub-280 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds003592-sub-280-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds003592-sub-280/bold[ses_1_task_rest_echo_1]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[ses_1_task_rest_echo_1]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds003592-sub-280 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds003592-sub-280-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds004173-sub-603309/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds004173-sub-603309 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004173-sub-603309-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds004173-sub-603309/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds004173-sub-603309 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004173-sub-603309-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds004173-sub-603309/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds004173-sub-603309 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004173-sub-603309-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds004884-sub-M2079/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds004884-sub-M2079 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004884-sub-M2079-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds004884-sub-M2079/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds004884-sub-M2079 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004884-sub-M2079-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds004884-sub-M2079/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds004884-sub-M2079 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004884-sub-M2079-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds004884-sub-M2079/bold[ses_217_task_naming40_acq_epfid2_dir_AP_run_2]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[ses_217_task_naming40_acq_epfid2_dir_AP_run_2]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds004884-sub-M2079 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004884-sub-M2079-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds004884-sub-M2079/bold[ses_217_task_rest_acq_epfidp2p2_dir_AP_run_9]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[ses_217_task_rest_acq_epfidp2p2_dir_AP_run_9]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds004884-sub-M2079 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004884-sub-M2079-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds004958-sub-fsm64ps/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds004958-sub-fsm64ps (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004958-sub-fsm64ps-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds004958-sub-fsm64ps/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds004958-sub-fsm64ps (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004958-sub-fsm64ps-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds004958-sub-fsm64ps/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds004958-sub-fsm64ps (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004958-sub-fsm64ps-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds004958-sub-fsm64ps/bold[task_rest]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_rest]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds004958-sub-fsm64ps (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds004958-sub-fsm64ps-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds005003-sub-RS001/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005003-sub-RS001 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005003-sub-RS001-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005003-sub-RS001/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005003-sub-RS001 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005003-sub-RS001-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005003-sub-RS001/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds005003-sub-RS001 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005003-sub-RS001-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds005003-sub-RS001/bold[ses_1_task_motor]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[ses_1_task_motor]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds005003-sub-RS001 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005003-sub-RS001-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds005003-sub-RS001/bold[ses_1_task_rest]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[ses_1_task_rest]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds005003-sub-RS001 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005003-sub-RS001-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds005040-sub-002/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005040-sub-002 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005040-sub-002-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005040-sub-002/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005040-sub-002 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005040-sub-002-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005040-sub-002/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds005040-sub-002 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005040-sub-002-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds005040-sub-002/bold[task_clip_run_1]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_clip_run_1]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds005040-sub-002 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005040-sub-002-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds005040-sub-002/bold[task_clip_run_1]/bold_fit/fmapreg/coregister | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_clip_run_1]/bold_fit/fmapreg/coregister` of the fMRIPrep run ds005040-sub-002 (bold_wf/bold_fit_wf/fmapreg_wf/coregister): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005040-sub-002-noreconall: fmapreg_wf/coregister>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds005454-sub-16/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005454-sub-16 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005454-sub-16-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005454-sub-16/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005454-sub-16 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005454-sub-16-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005454-sub-16/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds005454-sub-16 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005454-sub-16-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds005454-sub-16/bold[task_rest]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_rest]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds005454-sub-16 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005454-sub-16-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds005454-sub-16/bold[task_rest]/bold_fit/fmapreg/coregister | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[task_rest]/bold_fit/fmapreg/coregister` of the fMRIPrep run ds005454-sub-16 (bold_wf/bold_fit_wf/fmapreg_wf/coregister): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005454-sub-16-noreconall: fmapreg_wf/coregister>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
panel/fmriprep/ds005752-sub-ON01016/anat_fit/brain_extraction/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005752-sub-ON01016 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, full sizecommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005752-sub-ON01016-noreconall: brain_extraction_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005752-sub-ON01016/anat_fit/brain_extraction/norm/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/brain_extraction/norm` of the fMRIPrep run ds005752-sub-ON01016 (anat_fit_wf/brain_extraction_wf/norm): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005752-sub-ON01016-noreconall: brain_extraction_wf/norm, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in anat_to_template_Warped.nii.gz, standard output, transform0GenericAffine.mat, transform1InverseWarp.nii.gz …; identical to ANTs with CORE-MATH preloaded (every output) | 17/17 |
panel/fmriprep/ds005752-sub-ON01016/anat_fit/register_template[MNI152NLin2009cAsym]/registration/small | data panel: fMRIPrep's calls, every third voxel | fMRIPrep 25.2.6's antsRegistration node `anat_fit/register_template[MNI152NLin2009cAsym]/registration` of the fMRIPrep run ds005752-sub-ON01016 (anat_fit_wf/register_template_wf/registration): its command.txt replayed, its inputs from the run's working directory, on one voxel in 3 along each axis of its imagescommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005752-sub-ON01016-noreconall: _template_MNI152NLin2009cAsym/registration, every third voxel>' | Identical with correctly rounded maths differs from ANTs as built in ants_t1_to_mni_Warped.nii.gz, standard output, transformComposite.h5, transformInverseComposite.h5; identical to ANTs with CORE-MATH preloaded (every output) | 15/15 |
panel/fmriprep/ds005752-sub-ON01016/bold[ses_01_task_rest_dir_forward]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[ses_01_task_rest_dir_forward]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds005752-sub-ON01016 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005752-sub-ON01016-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Identical with correctly rounded maths differs from ANTs as built in standard output, transform0GenericAffine.mat; identical to ANTs with CORE-MATH preloaded (every output) | 11/11 |
panel/fmriprep/ds005752-sub-ON01016/bold[ses_01_task_rest_dir_reverse]/bold_fit/enhance_and_skullstrip_bold/norm | data panel: fMRIPrep's calls | fMRIPrep 25.2.6's antsRegistration node `bold[ses_01_task_rest_dir_reverse]/bold_fit/enhance_and_skullstrip_bold/norm` of the fMRIPrep run ds005752-sub-ON01016 (bold_wf/bold_fit_wf/enhance_and_skullstrip_bold_wf/norm): its command.txt replayed, its inputs from the run's working directorycommand lineantsRegistration '<fMRIPrep 25.2.6'"'"'s command.txt of ds005752-sub-ON01016-noreconall: enhance_and_skullstrip_bold_wf/norm>' | Bit-identical exit status, standard output, standard error, every file | 6/6 |
Speed
No benchmark report yet: the benchmark of antsRegistration is pending (next steps).
Random testing
Random testing runs random valid inputs and option combinations through the original and the replica, in long runs of several hours: inputs from seeded plain arithmetic (up to 32³ voxels; orientations, data types, scale factors and value ranges of every kind), with options drawn in random combinations from those the program accepts (how random testing works). A difference is shrunk to the smallest input and fewest options that still show it, then fixed and kept as a case of the record, registered as a defect of the original, or reported. The latest long run:
Random testing, the replica against the original: ANTs antsRegistration 2.6.5, the long run of 2026-10-25 (seed 20261025, 6.2 h for 21 programs): random valid inputs and option combinations, each run through both
| Program | Random cases | Extreme | Options covered | Agree | glibc | Notices | Undefined | Differ | Heavy | Guard | Unclear |
|---|---|---|---|---|---|---|---|---|---|---|---|
| antsRegistrationreplica | 2,110 | 0 | 114 | 1,802 | · | 60 | 164 | 74 | · | · | 10 |
- fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: exit status,ANTs with CORE-MATH: files written,ANTs with CORE-MATH: standard output: a Mattes stage whose second level’s analysis region (a sparse mask at full resolution) has one intensity: ITK keeps the joint histogram images of the first level, so SetSpacing’s exception for a bin size of 0 is not reached and the metric is 0; LARMORX threw the first level’s exception at every level in both modes. Now reproduce uses the bin size as ITK does; the default keeps the exception (rule B), both naming ITK-100 (larmorx_registration::metric::mattes::check_joint_pdf_spacing) - a defect of the original, in the register antsRegistration (replica), 5 case(s):
ANTs with CORE-MATH: files written,ANTs with CORE-MATH: standard error: the inverse composite of a list holding a displacement field without an inverse: ITK inverts it into a field transform with no parameters, whose HDF5 write fails (H5Pset_chunk) after writing part of the file; LARMORX writes no inverse composite and names ITK-96 in both modes; reproduce’s part-written file and HDF5’s messages are a follow-up - a defect of the original, in the register antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: files written,ANTs with CORE-MATH: standard error, thefixpass: the inverse composite of a list holding a displacement field without an inverse: ITK inverts it into a field transform with no parameters, whose HDF5 write fails (H5Pset_chunk) after writing part of the file; LARMORX writes no inverse composite and names ITK-96 in both modes; reproduce’s part-written file and HDF5’s messages are a follow-up - fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: rComposite.hN: a sampled metric with a shrink factor on a displacement-field stage: ITK adds the derivative of sample points outside the shrunk virtual region to other voxels’ parameters (ITK-97); reproduce now does the same and is bit-identical, the default drops them, both naming ITK-97 (larmorx_registration::metric::OutsideDerivatives) - fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: rComposite.hN,ANTs with CORE-MATH: standard output: ITK-97 reproduced: the derivatives of sample points outside the shrunk virtual region (a sampled B-spline displacement-field stage with a shrink factor) - a defect of the original, in the register antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: rNAffine.mat: without collapsing, an Affine stage whose Demons metric throws (ANTs goes on) adds no transform, so the next stage’s time-varying B-spline velocity field is written under the Affine stage’s .mat name (ANTS-51); there ANTs writes the transform object itself, TimeVaryingBSplineVelocityFieldTransform with its control-point lattice as the parameters, LARMORX the integrated displacement field. The default corrects ANTS-51 (each transform under its own stage’s name); reproducing the lattice under a .mat name is a follow-up - fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: rNBSpline.txt, thefixpass: a float B-spline transform parameter of exactly 0.173828125 lies halfway between two shortest decimals: ITK’s double-conversion writes the even one (0.17382812), Rust’s shortest formatting rounded up (0.17382813). LARMORX’s ITK number writer now takes the even one at an exact tie (larmorx_io::itk_transform::itk_number); the remaining fix-pass notice is ITK-31’s (no values changed) - fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: rNInverseWarp.nii.gz,ANTs with CORE-MATH: rNWarp.nii.gz: ITK-97 reproduced (as above): the derivatives of sample points outside the shrunk virtual region - fixed in LARMORX antsRegistration (replica), 2 case(s):
ANTs with CORE-MATH: rNInverseWarp.nii.gz,ANTs with CORE-MATH: rNWarp.nii.gz,ANTs with CORE-MATH: standard output,ANTs with CORE-MATH: w.nii.gz: ITK-97 reproduced (as above) - fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: rNInverseWarp.nii.gz,ANTs with CORE-MATH: rNWarp.nii.gz,ANTs with CORE-MATH: standard output, thefixpass: ITK-97 reproduced (as above) - fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: rNTranslation.mat: without collapsing (-z 0), a Translation stage whose Demons metric throws (it needs a displacement field; ANTs catches it and goes on) adds no transform and leaves ANTs’ names one short (ANTS-51), so the next stage’s displacement field is written under the linear stage’s .mat name; ANTs’ WriteTransform writes the stage’s own object there (GaussianSmoothingOnUpdateDisplacementFieldTransform), LARMORX a new DisplacementFieldTransform: the same parameters, another class name. Now the field keeps its stage’s class under a .mat or .txt name - fixed in LARMORX antsRegistration (replica), 3 case(s):
ANTs with CORE-MATH: rNWarp.nii.gz,ANTs with CORE-MATH: standard output: ITK-97 reproduced (as above) - fixed in LARMORX antsRegistration (replica), 5 case(s):
ANTs with CORE-MATH: rNWarp.nii.gz,ANTs with CORE-MATH: standard output,ANTs with CORE-MATH: w.nii.gz: a sampled metric (Random, 50 %) with a shrink factor of 2 on a B-spline displacement-field stage: ITK adds the derivative of sample points that fall outside the shrunk virtual domain to other voxels’ parameters (ITK-97); LARMORX dropped them in both modes, so the metric differed from the fourth iteration in the 12th digit. Now reproduce does as ITK does and is bit-identical; the default drops them; both name ITK-97 (larmorx_registration::metric::OutsideDerivatives) - fixed in LARMORX antsRegistration (replica), 26 case(s):
ANTs with CORE-MATH: standard error: ITK’s NIfTI writer warns when an image’s direction cosines are not orthogonal in float (an oblique quaternion that is orthogonal in double); LARMORX’s writer now prints the same warning (larmorx_io::itk_output) - fixed in LARMORX antsRegistration (replica), 4 case(s):
ANTs with CORE-MATH: standard error, thefixpass: the writer’s non-orthogonal direction warning (as above); the remaining notice is ITK-31’s (no values changed) - fixed in LARMORX antsRegistration (replica), 3 case(s):
ANTs with CORE-MATH: standard output: a stage whose sampled point set is empty: ITK’s exception is itkImageToImageMetricv4.hxx:263 with the metric’s own class name (MattesMutualInformationImageToImageMetricv4, …), or line 233 of the joint-histogram MI’s file; LARMORX printed line 268 and the base class - fixed in LARMORX antsRegistration (replica), 1 case(s):
ANTs with CORE-MATH: standard output, thefixpass: SyN and B-spline SyN take the iteration’s metric value as half the sum of the two halves’ values, the sum in the metric’s own type: with --float 1 two metrics with no valid point (FLT_MAX each) overflow to inf in ITK, and its convergence value follows; LARMORX summed in double. Now the sum is taken in the metric’s type - open: not yet fixed or explained (a clean room's difference is reported for its agent) antsRegistration (replica), 13 case(s): the
fixpass: the default’s exact floor changes a displacement field’s value in its last bit (antsRegistration’s field lookups, FieldXfm) and no notice names ITK-01 there: reproduce is exact; the notice in the field lookups is a follow-up - the comparison's own mistake, corrected antsRegistration (replica), 3 case(s):
exception: a 4D displacement-field composite (DisplacementFieldTransform_double_4_4), which lx.transforms.read does not hold: the comparison’s runner now reads its HDF5 parameters
Notices: a correction named in a case where it changed nothing the input shows (defect, cases): antsRegistration: ANTS-76 (1), ANTS-78 (12), ANTS-80 (3), ITK-30 (7), ITK-31 (17), ITK-42 (4), ITK-96 (25).
Every case is drawn from its tool, the run's seed and its number, and replays anywhere: how random testing works; each difference's triage: validation/difftest/triage.toml.
Known limits
- MINC transform files are refused with a message.
- The expected divergences: without a seed ANTs seeds its sampling from the clock (the replica uses a fixed seed), and so do three of the velocity-field stages whatever the seed (ANTS-82) and a linear stage after a displacement field (ANTS-72); crashes and reads of unwritten memory are not reproduced (ANTS-51, ANTS-66, ANTS-77); messages ANTs prints only with
--verbose(ANTS-45). - A stage after a B-spline transform runs in the replica (ITK-35, corrected), not in ANTs; give it a mask: the B-spline transform is the identity outside its valid region, so the composite jumps at that border.
- The comparison with ANTs is made on Linux x86-64; the cross-platform checks (how they work) hold the other five platforms to LARMORX’s own Linux output.
History
- 2026-10-21: the corrections of the point-set metrics, the B-spline field meshes and the stages after a B-spline transform decided, each measured (findings).
- 2026-10-27: the data panel’s record, every
antsRegistrationnode of the fourteen runs and the template registrations on every third voxel. - 2026-10-28: the standard comparison made again, with eight cases from random testing; the fourteen template registrations of the data panel at full size.
Data credits
The data this page shows or names: each dataset with its version, DOI and licence, from LARMORX's test-data catalog. Public-domain dedications (CC0, PDDL) ask nothing; the datasets are cited all the same.
- Mixed-gambles task (OpenNeuro ds000005, version 00001), Open Data Commons Public Domain Dedication and Licence 1.0.
- Children and adults watching a short film (OpenNeuro ds000228, version 1.1.1, doi:10.18112/openneuro.ds000228.v1.1.1), CC0 1.0 public-domain dedication.
- Behavioral interventions for reducing head motion in children (OpenNeuro ds000256, version 00002), Open Data Commons Public Domain Dedication and Licence 1.0.
- Example fieldmaps (OpenNeuro ds001600, version 1.0.0), Creative Commons Attribution-ShareAlike (CC BY-SA; the dataset gives no version). Attribution: Cieslak M., Elliott M., Satterthwaite T. Example Fieldmaps. OpenNeuro ds001600, version 1.0.0 (Siemens Prisma, University of Pennsylvania). The images shown are adapted from it: reoriented to RAS, cropped to the region shown, masked outside it, and shown as the outputs of the programs compared. Share-alike: the images adapted from this dataset are shared under the same licence, CC BY-SA.
- fMRI: resting state and arithmetic task (OpenNeuro ds002422, version 1.1.0, doi:10.18112/openneuro.ds002422.v1.1.0), CC0 1.0 public-domain dedication.
- MB-EPI-comparison (OpenNeuro ds002603, version 1.0.0, doi:10.18112/openneuro.ds002603.v1.0.0), CC0 1.0 public-domain dedication.
- Neurocognitive aging data release (OpenNeuro ds003592, version 1.0.13, doi:10.18112/openneuro.ds003592.v1.0.13), CC0 1.0 public-domain dedication.
- Movement-related artefacts (MR-ART) (OpenNeuro ds004173, version 1.0.2, doi:10.18112/openneuro.ds004173.v1.0.2), CC0 1.0 public-domain dedication.
- Aphasia Recovery Cohort (OpenNeuro ds004884, version 1.0.2, doi:10.18112/openneuro.ds004884.v1.0.2), CC0 1.0 public-domain dedication.
- The FreeSurfer Maintenance Dataset (OpenNeuro ds004958, version 1.0.0, doi:10.18112/openneuro.ds004958.v1.0.0), CC0 1.0 public-domain dedication.
- Retrospective Task/Rest fMRI Data from Tumor Patients (OpenNeuro ds005003, version 2.0.0, doi:10.18112/openneuro.ds005003.v2.0.0), CC0 1.0 public-domain dedication.
- Political Moralization (OpenNeuro ds005040, version 1.2.0, doi:10.18112/openneuro.ds005040.v1.2.0), CC0 1.0 public-domain dedication.
- fMRIPrep boot camp 7T training dataset (OpenNeuro ds005454, version 1.0.0, doi:10.18112/openneuro.ds005454.v1.0.0), Creative Commons Attribution 4.0 (CC BY 4.0). Attribution: Sleight EC, Esteban O, Grouiller F. fMRIPrep boot camp -- 7T training dataset. OpenNeuro ds005454, version 1.0.0. doi:10.18112/openneuro.ds005454.v1.0.0 The images shown are adapted from it: reoriented to RAS, cropped to the region shown, masked outside it, and shown as the outputs of the programs compared.
- NIMH Healthy Research Volunteer Dataset (OpenNeuro ds005752, version 2.1.0, doi:10.18112/openneuro.ds005752.v2.1.0), CC0 1.0 public-domain dedication.
- MNI152NLin2009cAsym template (TemplateFlow, version tpl-MNI152NLin2009cAsym git 15d7c02160f79f5218d2545b4febebeecc11531d), the MNI ICBM 152 copyright notice.