NiftyReg

reg_f3d with its defaults (cubic B-spline, NMI), from the corpus's rigidly aligned pair.

RegistrationCPUversion v2.0.0 (ee6a92d)every datasetBSD-3-Clause

Run it

With its KonfAI app, at the benchmark's settings (the fidopt app repositories are private for now: a Hugging Face token with access opens them):

pip install konfai-apps
konfai-apps infer fideus-labs/fidopt-registration:NiftyReg -i fixed.nii.gz -i moving.nii.gz -o out

The app on Hugging Face: its model, its settings and how it is scored.

Where it works

Its place on each kind of problem it has results on, among the methods ranked there. A method that leads one problem can trail another: read the one closest to yours.

Problem	Place	Ranked	Datasets
Heart CT, same animal, months apart	1	of 1	1 of 1
Microscopy (2D), two modalities, known deformation	1	of 1	1 of 1
Rigid registration, CT to ultrasound	1	of 3	1 of 1
Repeat chest CT, same patient, tumour truth	2	of 5	1 of 1
Temporal bone CBCT, between donors	2	of 4	1 of 1
Repeat scans, same patient	4	of 5	1 of 1
Simulated thoracic CBCT, known motion	4	of 4	1 of 1
Abdomen CT, same patient		not ranked yet	1 of 3
Brain MR, inter-subject		not ranked yet	1 of 4
Two modalities, same patient		not ranked yet	2 of 5
Head, neck and spine, inter-patient		not ranked yet	2 of 4
Abdomen MR, inter-patient		not ranked yet	1 of 2
Radiographs (2D), inter-patient		not ranked yet	1 of 2
Known deformation, two images of one session		not ranked yet	2 of 5
Same patient, truth made by a program		not ranked yet	3 of 8

Results

Dataset	#	Read by	Score	Change	SDlogJ	Time s
Aariz cephalograms	7	TRE mm	11.93	−0.67	3.219	19
Abdominal-DIR-QA	3	Dice	0.775	+0.344	0.479	174
OpenEar, inter-specimen temporal-bone CBCT	2	Dice	0.469	−0.014	2.190	773
ORION colorectal cancer, H&E and Hoechst of one section, known warp	1	TRE µm	9.71	−0.98	1.376	19
PROMISE12	4	Dice	0.633	+0.017	0.368	21
PSMA-PET-CT-Lesions follow-up (Learn2Reg 2026 PSMAReg)	1	TRE mm	5.39	−2.20	0.075	308
Pelvic-Reference-Data	3	TRE mm	4.85	+2.13	0.127	97
Pelvic landmarks CT/CT (LUMC)	3	Dice	0.731	+0.116	0.026	189
Pelvic landmarks MR/CT (LUMC)	3	Dice	0.597	+0.067	0.103	29
Cardiac CT of growing pigs (A heart for pigs)	1	Dice	0.875	+0.161	2.151	750
Prostate-MRI-US-Biopsy	3	Dice	0.299	−0.320	1.458	19
LCTSC CT and simulated CBCT, known deformation	4	TRE mm	0.78	−0.46	0.050	24
QIN-PROSTATE-Repeatability	4	Dice	0.626	−0.017	0.365	46
QUADRA_HC, whole-body CT test-retest	1	Dice	0.840	+0.288	0.030	316
RIDER Lung CT	2	Dice	0.785	+0.049	0.061	293
RibSeg v2	2	Dice	0.312	+0.269	0.774	458
SHG/BF TMA cores (Eliceiri)	1	TRE px	22.72	+1.35	0.467	2
SPIDER	1	Dice	0.829	+0.244	0.441	35
Spine-generic multi-subject	1	Dice	0.853	+0.524	0.183	30
Spine CT and ultrasound, two ex-vivo phantoms	1	TRE mm	16.05	−178.53	0.000	39
SynthRAD2023 CBCT/CT, brain11/18 pairs		Dice	0.740	+0.076	0.036	48
SynthRAD2023 MR/CT, brain8/18 pairs		Dice	0.496	−0.078	0.074	45
SynthRAD2023 CBCT/CT, pelvis8/18 pairs		Dice	0.410	−0.035	0.036	35
SynthRAD2023 MR/CT, pelvis	1	Dice	0.699	+0.004	0.069	63
Tongue musculature MRI	4	Dice	0.607	+0.109	0.034	44
STS-3D-Tooth	1	Dice	0.711	+0.266	1.451	486
TopBrain 2025/2026	2	Dice	0.604	+0.156	0.594	228
TopCoW 2024	2	Dice	0.746	+0.135	0.426	203
TotalSegmentator MRI	4	Dice	0.268	+0.084	0.247	16
TrackRAD2025, two frames of a sagittal cine MR sequence	4	Dice	0.711	+0.176	0.060	1
UCSD-BMETS-Longitudinal	1	Dice	0.768	−0.023	0.038	270
UVDoc, measured paper geometry with rendered document texture	7	TRE px	368.98	−0.54	0.185	12
UltraCortex	3	Dice	0.636	+0.138	0.205	150
Vestibular-Schwannoma-SEG	2	Dice	0.565	+0.310	0.081	62
WxBS	2	TRE px	176.86	−1.62	0.491	3

Source: github.com/KCL-BMEIS/niftyreg
Reference: Modat et al., Computer Methods and Programs in Biomedicine 2010