Pith. sign in

REVIEW 3 major objections 8 minor 1 cited by

Beyond the LUMIR challenge: The pathway to foundational registration models

T0 review · 3 major / 8 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The LUMIR challenge shows that unsupervised deep learning models, trained solely on 3,384 T1-weighted brain MRIs, outperform optimization-based registration methods and generalize zero-shot across disease populations, imaging protocols…

desk verdict Solid challenge benchmark with a genuinely useful dataset, but the abstract overclaims zero-shot robustness and the main ranking ignores its own Bonferroni correction. read the letter →

arxiv 2505.24160 v3 pith:QFNDA3DV submitted 2025-05-30 eess.IV cs.CV

Junyu Chen , Shuwen Wei , Joel Honkamaa , Pekka Marttinen , Hang Zhang , Min Liu , Yichao Zhou , Zuopeng Tan
show 28 more authors
Zhuoyuan Wang Yi Wang Hongchao Zhou Shunbo Hu Yi Zhang Qian Tao Lukas Förner Thomas Wendler Bailiang Jian Benedikt Wiestler Tim Hable Jin Kim Dan Ruan Frederic Madesta Thilo Sentker Wiebke Heyer Lianrui Zuo Yuwei Dai Jing Wu Jerry L. Prince Harrison Bai Yong Du Yihao Liu Alessa Hering Reuben Dorent Lasse Hansen Mattias P. Heinrich Aaron Carass
This is my paper
classification eess.IVcs.CV MSC 68T0792C55
keywords imageregistrationbrainMRIunsupervisedlearningzero-shotgeneralizationmedicalimagingbenchmarkdeepdiffeomorphicdeformationfoundationmodels
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The LUMIR challenge paper reports that unsupervised deep-learning registration models, trained without any anatomical labels on 3,384 T1-weighted brain MRIs from healthy controls, outperform leading optimization-based methods on both the in-domain test set and extensive zero-shot tasks, while producing smooth, near-diffeomorphic deformation fields. The zero-shot evaluations cover pediatric ADHD scans, Alzheimer's disease cohorts at 1.5T and 3T, an independent T1-weighted dataset, 9.4T ultra-high-field images, T2-weighted, T2*-weighted, and FLAIR contrasts, and macaque brains. The result matters because it suggests that large-scale data plus a consistent preprocessing pipeline may be sufficient for learned models to act as general-purpose intra-contrast brain registration tools, directly challenging the concern that deep-learning registration breaks under domain shift. The authors are explicit that this generalization is conditional on all data going through the same skull-stripping, affine normalization, intensity normalization, resampling, and cropping as the training set.

What carries the argument

The load-bearing object is the LUMIR challenge protocol itself: 3,384 unlabeled T1-weighted brain MRIs drawn from 64 sites, every image processed identically by skull-stripping, affine alignment to a 1 mm MNI template, kernel-based intensity normalization, resampling, and cropping, giving over 5.7 million possible training pairs, and evaluated on 590 in-domain images plus zero-shot datasets run through the same pipeline. Within that protocol, the architectural patterns that recur among the top performers are dual-stream encoders that extract features from moving and fixed images separately, coarse-to-fine deformation estimation across resolutions, and progressive registration that iteratively warps the moving image toward the fixed image; the very best methods are single-step at inference, and inverse consistency by construction appears in several top entries, while instance-specific optimization helps robustness but is not needed for top accuracy.

What would settle it

Register the NIMH T1-weighted and FLAIR zero-shot images with the top-performing model after resampling them to 2 mm isotropic instead of 1 mm, keeping everything else identical; the paper predicts substantial accuracy degradation because deep networks are sensitive to resolution changes in their receptive fields. If the model's Dice scores and target registration error stay at the 1 mm levels, the claim that generalization requires a consistent preprocessing pipeline would be falsified.

Watch

Extended reading notes

Core claim

On the challenge's own terms, the central discovery is that deep-learning registration networks trained purely on 3,384 unlabeled T1-weighted brain MRIs achieve the highest registration accuracy on a private 590-image test set (mean Dice around 0.78, mean target registration error around 3.1 mm) and remain the top performers on most zero-shot out-of-domain tasks, including unseen MRI contrasts, pathological populations, ultra-high-field acquisitions, and macaque brains. The top methods also generate anatomically plausible deformations, with non-diffeomorphic volume below roughly 1% of the brain, whereas the best traditional optimization-based method places 12th out of 21 in-domain. The paper reads this as evidence that, with sufficient training data and an identical preprocessing pipeline, learning-based models can match or surpass modality-agnostic alternatives on intra-contrast tasks; the clear boundary is cross-contrast generalization, where all methods, including the best deep models, show substantial drops on T2*-weighted and FLAIR images.

Load-bearing premise

The zero-shot evaluations assume that every out-of-domain image goes through exactly the same preprocessing pipeline as the training data — skull-stripping, affine alignment to the 1 mm MNI template, intensity normalization, resampling, and cropping — so the claimed generalization is conditional on pipeline consistency, not an inherent property of the models.

Editorial extensions

If this is right

  • If the reported zero-shot results are correct, a model trained on a few thousand T1-weighted brain scans can be dropped onto a new human brain dataset, run through the same preprocessing, and register it without any retraining or per-pair optimization.
  • The same protocol could become a reusable evaluation harness: any new registration architecture can be tested for both in-domain accuracy and zero-shot robustness using the LUMIR preprocessing and ranking pipeline.
  • The observed design correlates (dual-stream encoders, coarse-to-fine estimation, progressive registration) give concrete architectural guidance for building future registration models, even though the paper does not run controlled ablations.
  • The clear degradation on T2*-weighted and FLAIR data marks the current limit of T1-only foundation-model claims, so a genuinely general registration model must close the unseen-contrast gap rather than rely on intra-contrast robustness.
  • Optimization-based methods still have a deployment role where preprocessing cannot be standardized, since they can be tuned per instance, but they no longer define the accuracy frontier for this task.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension the paper leaves implicit is deliberately perturbing the preprocessing (voxel size, cropping, skull-stripping) on the zero-shot sets; the paper's own conditionality predicts a large accuracy drop, and quantifying that drop would turn a stated limitation into a measured robustness curve.
  • If the scale-dependence is causal, the same recipe — large unlabeled single-contrast data plus a fixed preprocessing container — may transfer to other organs or modalities, potentially making contrast-agnostic training strategies less necessary than previously thought.
  • Deployment in practice would likely require shipping the exact preprocessing steps alongside the trained model, because the paper's claims do not extend to arbitrary pipelines; this is a practical corollary of the stated conditionality rather than a claim the paper makes.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 8 minor

Summary. This manuscript presents the Large-scale Unsupervised Brain MRI Image Registration (LUMIR) challenge and its follow-up analysis. The challenge provides 4,014 T1-weighted brain MRIs (3,384 for training) assembled from OpenBHB and AFIDs-OASIS, a held-out test set of 590 subjects with SLANT parcellation labels on 460 images and manual anatomical landmarks on 130 images, and a ranking protocol based on DSC, HD95, TRE, NDV, DSC30, and TRE30. Beyond the in-domain evaluation, the paper reports zero-shot results on six out-of-domain settings — ADHD children, ADNI 1.5T and 3T, NIMH T1-w/T2-w/T2*-w/FLAIR, UltraCortex 9.4T, and macaque brains — covering inter-subject, atlas-to-subject, and subject-to-atlas tasks. Eleven participant submissions and ten baselines (ANTs SyN, deedsBCV, FireANTs, SynthMorph, TransMorph, uniGradICON, VFA, VoxelMorph) are compared. The central claims are that deep learning methods dominated both in-domain and zero-shot rankings, that the best methods produced smooth, near-diffeomorphic deformation fields, that they outperformed optimization-based methods, and that large-scale T1-w training combined with a consistent preprocessing pipeline enables robust generalization, suggesting a pathway to foundational registration models. The paper also analyzes inverse consistency, the DSC-TRE correlation, deformation regularity, and architectural factors such as dual-stream encoders and progressive registration.

Significance. The challenge resource is a significant contribution: it is the largest public T1-w brain MRI collection assembled specifically for unsupervised registration training, and the evaluation is unusually thorough, combining label-based (DSC, HD95), landmark-based (TRE), and deformation-regularity (NDV) metrics, with means, standard deviations, rankings, and Bonferroni-corrected significance matrices in the appendix. The landmark-based evaluation is genuinely orthogonal to label overlap, and the zero-shot program is ambitious, with honest reporting of negative results (contrast-specific degradation, SynthMorph winning the T2*-w task, FireANTs competitive on macaque). The discussion explicitly acknowledges the preprocessing conditionality, the use of automated labels, and the absence of severe pathology and inter-contrast tasks. With the qualifiers the authors state in Sections 6.1 and 6.2, the paper provides a valuable benchmark and a credible empirical map of where T1-w-trained unsupervised registration does and does not transfer. The main shortcoming is that the abstract and introduction state the findings unconditionally, making the advertised conclusion broader than the evidence.

major comments (3)
  1. [Abstract and Section 1; cf. Sections 6.1 and 6.2] The abstract asserts that deep learning methods "remained robust to most domain shifts" and have "potential to serve as a foundation model for general-purpose medical image registration," and Section 1 claims "excellent out-of-domain generalization... even surpassing modality-agnostic models such as SynthMorph." These statements are unconditional, yet the paper's own discussion heavily qualifies them: Section 6.1 states that "the generalization observed in this study should be understood as conditional on consistent preprocessing rather than as an inherent property of the learned models," and Section 6.2 warns that under different resolutions, sampling schemes, or input sizes "learning-based methods may struggle to operate out of the box." A foundation-model claim is precisely a promise of out-of-the-box operation on new data, so the headline claims exceed the evidence, which supports the narrower conclusion that LUMIR-trained models generalize within a matched pipeline (skull-stripping, MNI affine alignment, 1 mm resampling, intensity normalization) to unseen populations, most contrasts, and, with a manual scaling step, to macaques. The abstract and introduction should carry the same qualifier that the discussion carries.
  2. [Section 1; Appendix B.1.4, Tables B.14 and B.16] The claim that LUMIR-trained models surpass the contrast-agnostic SynthMorph is not uniformly supported by the per-contrast zero-shot results. On NIMH T2*-w inter-subject registration, SynthMorph ranks 1st (Table B.14: DSC 0.656, HD95 3.703), ahead of all LUMIR-trained methods (MadeForLife 2nd, honkamj 7th); on FLAIR, SynthMorph ranks 2nd (Table B.16) behind MadeForLife. Section 5.2 itself notes "a noticeable drop in performance when evaluated on unseen contrasts, especially T2*-w and FLAIR images." The claim should be restricted to the contrasts and tasks where it holds (e.g., T1-w, T2-w, and the atlas-based tasks) or rephrased to state that large-scale T1-w training reduces but does not eliminate the advantage of contrast-agnostic training.
  3. [Section 2.2; Tables 2 and 3] The primary ranking uses one-sided Wilcoxon signed-rank tests at per-comparison α=0.05 with no adjustment for multiple comparisons across 21 methods, a choice inherited from Learn2Reg but not without consequence here: the top test-set methods differ by ΔDSC ≈ 0.007 (0.785 vs 0.778 vs 0.778, Table 2), and the fine-grained ordering feeds the medal table (Table 3) and the top-three qualitative figures. The class-level conclusions (deep learning over optimization-based; near-diffeomorphic fields) do not depend on this ordering, but the paper should either hedge the specific rankings or report whether the ordering survives the Bonferroni-corrected pairwise tests already presented in Appendix C.
minor comments (8)
  1. [Abstract; Section 2.1.1] The abstract states that LUMIR provides "4,014 unlabeled T1-weighted MRIs for training," but Section 2.1.1 allocates only 3,384 images to the training partition; 4,014 is the total dataset size including the validation and test splits.
  2. [Section 2.2; Section 6.8] NDV is defined in Section 2.2 as a volumetric measure "in units of voxels," while Section 6.8 defines it as "the proportion of folded volume relative to the total brain volume"; Table 2 values (e.g., 2.52e-03) are consistent with a proportion, so the two definitions should be reconciled.
  3. [Section 6.1; Appendix B.1.4] Section 6.1 states that "the relative ranking of the top-performing deep learning methods remained stable" across zero-shot tasks, whereas Appendix B.1.4 describes "chaotic changes in the ranking" between the T1-w and T2*/FLAIR NIMH tasks; these statements should be reconciled by specifying whether they refer to the method class or to individual ranks.
  4. [References; Section 6.7] The reference list contains two near-identical entries for Rohlfing (2011 and 2012), both titled "Image similarity and tissue overlaps as surrogates for image registration accuracy: Widely used but unreliable" with the same journal and page numbers; the duplicate should be removed and the citations in Sections 2.2 and 6.7 aligned.
  5. [Section 3; Section 4.1] Section 3 reports that 12 teams submitted trained models for the test phase, while Section 4.1 says the evaluation received submissions from 11 teams; the discrepancy should be explained.
  6. [Abstract; Section 5.1] The abstract states that deep learning methods "produced anatomically plausible, diffeomorphic deformation fields," but Section 5.1 describes the top methods as "approaching diffeomorphic quality" with NDV well below 1% rather than exactly zero; the abstract wording should match this nuance.
  7. [Section 2.1.2; Appendix B.3.6] The macaque task applies a manually chosen 1.5× upscaling factor and a qualitatively verified affine alignment to the MNI template; since this is the only cross-species task and its results are summarized in Section 5.2, the paper should state that the scaling factor was not varied and may affect cross-species conclusions.
  8. [Various] Typos and typesetting issues: "Haudorff" (Appendix B.1.4), "the onther three contrasts" (Appendix B.1.4), "during raining" (Section 4.2, SynthMorph entry), "run the gambit" (Section 6.7), and a malformed inverse-consistency identity in Section 6.6.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central results rest on held-out labels, landmarks, and zero-shot evaluations rather than on fitted parameters or self-referential definitions.

full rationale

The paper's core claims are empirical benchmark results. The in-domain test set comprises 590 privately held subjects with SLANT label maps and 130 landmark-annotated images; the zero-shot datasets (ADHD, ADNI, NIMH, UltraCortex, Macaque) are explicitly non-overlapping with the training data. All accuracy metrics (DSC, HD95, TRE, DSC30, TRE30) and the deformation-regularity metric (NDV) are computed on these held-out annotations, so no reported score is a fitted parameter renamed as a prediction. Baselines are either training-free optimization methods (ANTsSyN, deedsBCV, FireANTs), pretrained external foundation models used with their published weights (SynthMorph, uniGradICON), or LUMIR-trained baselines (TransMorph, VFA, VoxelMorph); none of these make the evaluation self-referential. The 'foundation model' generalization claim is explicitly qualified in Sec. 6.1 and Sec. 6.2 as conditional on the same preprocessing pipeline being applied to zero-shot data, and the paper itself acknowledges that performance is expected to degrade under different resolutions, sampling schemes, or input sizes. That mismatch between the abstract's unqualified wording and the discussion's caveat is an overstatement or correctness concern, not circularity. The presence of organizer-participants and self-citations (e.g., Chen et al. 2025, Liu et al. 2024a, Honkamaa and Marttinen 2024) is normal in a challenge summary paper and is not load-bearing: the main conclusions are supported by the held-out evaluation data themselves, not by the cited prior work. No equation in the paper defines one quantity in terms of another such that a 'prediction' equals its own input by construction.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The paper's central empirical claims rest on domain assumptions about diffeomorphic deformations, SLANT label reliability, and matched preprocessing across zero-shot data. The only hand-selected numeric constant affecting a specific evaluation is the macaque upscaling factor. There are no invented entities, and the training parameters of each method are not the paper's own fitted quantities.

free parameters (1)
  • Macaque brain upscaling factor = 1.5
    Hand-chosen factor to resize macaque brains toward human MNI scale before evaluation; it influences all cross-species zero-shot results (Sec. 2.1.2).
assumptions (3)
  • domain assumption Brain deformation between healthy and neurodegenerative subjects is diffeomorphic and topology-preserving
    Used to justify NDV as a validity metric and to exclude pathologies with large topological changes (Sec. 6.8 and Sec. 6.1).
  • domain assumption SLANT automatically generated 133-structure labels are accurate enough to serve as an evaluation gold standard
    DSC and HD95 rankings rely on these labels; the paper acknowledges segmentation errors can influence scores (Sec. 6.9).
  • domain assumption Intensity normalization and affine alignment make all zero-shot datasets comparable to LUMIR training data
    Zero-shot generalization is measured only under this matched preprocessing condition (Sec. 2.1.2 and Sec. 6.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Beyond the LUMIR challenge: The pathway to foundational registration models." pith.science (2026). https://pith.science/paper/QFNDA3DV

@misc{pith2026250524160,
  author       = {Pith},
  title        = {Pith review of: Beyond the LUMIR challenge: The pathway to foundational registration models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QFNDA3DV}},
  note         = {Machine review of arXiv:2505.24160}
}
read the original abstract

Medical image challenges have played a transformative role in advancing the field, catalyzing innovation and establishing new performance benchmarks. Image registration, a foundational task in neuroimaging, has similarly advanced through the Learn2Reg initiative. Building on this, we introduce the Large-scale Unsupervised Brain MRI Image Registration (LUMIR) challenge, a next-generation benchmark for unsupervised brain MRI registration. Previous challenges relied upon anatomical label maps, however LUMIR provides 4,014 unlabeled T1-weighted MRIs for training, encouraging biologically plausible deformation modeling through self-supervision. Evaluation includes 590 in-domain test subjects and extensive zero-shot tasks across disease populations, imaging protocols, and species. Deep learning methods consistently achieved state-of-the-art performance and produced anatomically plausible, diffeomorphic deformation fields. They outperformed several leading optimization-based methods and remained robust to most domain shifts. These findings highlight the growing maturity of deep learning in neuroimaging registration and its potential to serve as a foundation model for general-purpose medical image registration.

Figures

Figures reproduced from arXiv: 2505.24160 by the authors.

Figure 1
Figure 1. The top panel provides a breakdown of the training, validation, and testing data used in the LUMIR Challenge; with each source listed and the [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Representative registration results from the challenge evaluation task. Shown are the moving and fixed images, the deformation fields and deformed [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Representative registration results from the zero-shot evaluation tasks. From top to bottom the three panels show the Intersubject, Atlas to [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Shown for each subtask is a ranking for registration accuracy (ACC), 30 [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Surrogate Supervision for Robust and Generalizable Deformable Image Registration

    cs.CV 2025-09 conditional novelty 5.0 of 10

    Surrogate supervision applies the registration loss to clean surrogate images rather than raw inputs, improving robustness to artifacts, masks, and modality differences without extra inference cost.

Reference graph

Works this paper leans on

92 extracted references · 76 canonical work pages · cited by 1 Pith paper

  1. [1]

    Murphy, B

    K. Murphy, B. van Ginneken, J. M. Reinhardt, S. Kabus, K. Ding, X. Deng, K. Cao, K. Du, G. E. Christensen, V. Garcia, T. Vercauteren, N. Ayache, O. Commowick, G. Malandain, B. Glocker, N. Paragios, N. Navab, V. Gorbunova, J. Sporring, M. de Bruijne , X. Han, M. P. Heinrich, J. A. Schnabel, M. Jenkinson, C. Lorenz, M. Modat, J. R. McClelland, S. Ourselin, ...

    1901

  2. [2]

    Sirinukunwattana, J

    K. Sirinukunwattana, J. P. W. Pluim, H. Chen, X. Qi, P.-A. Heng, Y. B. Guo, L. Y. Wang, B. J. Matuszewski, E. Bruni, U. Sanchez, A. B\" o hm, O. Ronneberger, B. B. Cheikh, D. Racoceanu, P. Kainz, M. Pfeiffer, M. Urschler, D. R. J. Snead, and N. M. Rajpoot, `` Gland Segmentation in Colon Histology Images: The GlaS Challenge Contest ,'' Medical Image Analys...

    2017

  3. [3]

    a rz, O. Maier, K. Maier-Hein, B. H. Menze, H. M\

    L. Maier-Hein, M. Eisenmann, A. Reinke, S. Onogur, M. Stankovic, P. Scholz, T. Arbel, H. Bogunovic, A. P. Bradley, A. Carass, C. Feldmann, A. F. Frangi, P. M. Full, B. van Ginneken , A. Hanbury, K. Honauer, M. Kozubek, B. A. Landman, K. M\" a rz, O. Maier, K. Maier-Hein, B. H. Menze, H. M\" u ller, P. F. Neher, W. Niessen, N. Rajpoot, G. C. Sharp, K. Siri...

    2018

  4. [4]

    S. S. Halabi, L. M. Prevedello, J. Kalpathy-Cramer, A. B. Mamonov, A. Bilbily, M. Cicero, I. Pan, L. A. Pereira, R. T. Sousa, N. Abdala, F. C. Kitamura, H. H. Thodberg, L. Chen, G. Shih, K. Andriole, M. D. Kohli, B. J. Erickson, and A. E. Flanders, `` The RSNA pediatric bone age machine learning challenge ,'' Radiology, vol. 290, no. 2, pp. 498--503, 2019

    2019

  5. [5]

    Antonelli, A

    M. Antonelli, A. Reinke, S. Bakas, K. Farahani, A. Kopp-Schneider, B. A. Landman, G. Litjens, B. Menze, O. Ronneberger, R. M. Summers, B. van Ginneken11 , M. Bilello, P. Bilic, P. F. Christ, R. K. G. Do, M. J. Gollub, S. H. Heckers, H. Huisman, W. R. Jarnagin, M. K. McHugo, S. Napel, J. S. G. Pernicka, K. Rhode1, C. Tobon-Gomez, E. Vorontsov, J. A. Meakin...

    2022

  6. [6]

    o lund, D. Grzech, H. Qiu, Z. Li, A. Thorley, J. Duan, C. Gro br\

    A. Hering, L. Hansen, T. C. W. Mok, A. C. S. Chung, H. Siebert, S. H\" a ger, A. Lange, S. Kuckertz, S. Heldmann, W. Shao, S. Vesal, M. Rusu, G. Sonn, T. Estienne, M. Vakalopoulou, L. Han, Y. Huang, P.-T. Yap, M. Brudfors, Y. Balbastre, S. Joutard, M. Modat, G. Lifshitz, D. Raviv, J. Lv, Q. Li, V. Jaouen, D. Visvikis, C. Fourcade, M. Rubeaux, W. Pan, Z. X...

    2023

  7. [7]

    B. H. Menze, A. Jakab, S. Bauer, J. Kalpathy-Cramer, K. Farahani, J. Kirby, Y. Burren, N. Porz, J. Slotboom, R. Wiest, L. Lanczi, E. Gerstner, M.-A. Weber, T. Arbel, B. B. Avants, N. Ayache, P. Buendia, D. L. Collins, N. Cordier, J. J. Corso, A. Criminisi, T. Das, H. Delingette, C . Demiralp, C. R. Durst, M. Dojat, S. Doyle, J. Festa, F. Forbes, E. Geremi...

    1993

  8. [8]

    a ni, M. P. Heinrich, M. Liebrand, S. Winzeck, A. Basit, P. Bentley, L. Chen, D. Christiaens, F. Dutil, K. Egger, C. Feng, B. Glocker, M. G\

    O. Maier, B. H. Menze, J. von der Gablentz , L. H\" a ni, M. P. Heinrich, M. Liebrand, S. Winzeck, A. Basit, P. Bentley, L. Chen, D. Christiaens, F. Dutil, K. Egger, C. Feng, B. Glocker, M. G\" o tz, T. Haeck, H.-L. Halme, M. Havaei, K. M. Iftekharuddin, P.-M. Jodoin, K. Kamnitsas, E. Kellner, A. Korvenoja, H. Larochelle, C. Ledig, J.-H. Lee, F. Maes, Q. ...

    2015

Show all 92 references
  1. [9]

    Carass, S

    A. Carass, S. Roy, A. Jog, J. L. Cuzzocreo, E. Magrath, A. Gherman, J. Button, J. Nguyen, F. Prados, C. H. Sudre, M. J. Cardoso, N. Cawley, O. Ciccarelli, C. A. M. Wheeler-Kingshott, S. Ourselin, L. Catanese, H. Deshpande, P. Maurel, O. Commowick, C. Barillot, X. Tomas-Fernand...

    2017

  2. [10]

    Commowick, A

    O. Commowick, A. Istace, M. Kain, B. Laurent, F. Leray, M. Simon, S. C. Pop, P. Girard, R. Ameli, J. C. Ferr\' e , A. Kerbrat, T. Tourdias, F. Cervenansky, T. Glatard, J. Beaumont, S. Doyle, F. Forbes, J. Knight, A. Khademi, A. Mahbod, C. Wang, R. Mckinley, F. Wagner, J. Musch...

    2018

  3. [11]

    H. J. Kuijf, J. M. Biesbroek, J. de Bresser , R. Heinen, S. Andermatt, M. Bento, M. Berseth, M. Belyaev, M. J. Cardoso, A. Casamitjana, D. L. Collins, M. Dadar, A. Georgiou, M. Ghafoorian, D. Jin, A. Khademi, J. Knight, H. Li, X. Llad\' o , M. Luna, Q. Mahmood, R. McKinley, A....

    2019

  4. [12]

    A. M. Mendrik, K. L. Vincken, H. J. Kuijf, M. Breeuwer, W. Bouvy, J. de Bresser , A. Alansary, M. de Bruijn , A. Carass, A. El-Baz, A. Jog, R. Katyali, A. R. Khan, F. van der Lijn , Q. Mahmood, R. Mukherjee, A. van Opbroek , S. Paneri, S. Pereira, M. Persson, M. Rajchl, D. Sar...

    2015

  5. [13]

    Carass, J

    A. Carass, J. L. Cuzzocreo, S. Han, C. R. Hernandez-Castillo, P. E. Rasser, M. Ganz, V. Beliveaux, J. Dolz, I. B. Ayed, C. Desrosiers, B. Thyreau, J. E. Romero, P. Coup\' e , J. V. Manj\' o n, V. S. Fonov, D. L. Collins, S. H. Ying, D. Crocetti, B. A. Landman, S. H. Mostofsky,...

    2018

  6. [14]

    Rohlfing, `` Image similarity and tissue overlaps as surrogates for image registration accuracy: Widely used but unreliable ,'' IEEE Trans

    T. Rohlfing, `` Image similarity and tissue overlaps as surrogates for image registration accuracy: Widely used but unreliable ,'' IEEE Trans. Med. Imag., vol. 31, no. 2, pp. 153--163, 2011

    2011

  7. [15]

    J. Chen, Y. Liu, S. Wei, Z. Bian, S. Subramanian, A. Carass, J. L. Prince, and Y. Du, `` A survey on deep learning in medical image registration: New technologies, uncertainty, evaluation metrics, and beyond ,'' Medical Image Analysis, p. 103385, 2024

    2024

  8. [16]

    Hoffmann, B

    M. Hoffmann, B. Billot, D. N. Greve, J. E. Iglesias, B. Fischl, and A. V. Dalca, ``Synthmorph: learning contrast-invariant registration without acquired images,'' IEEE Trans. Med. Imag., vol. 41, no. 3, pp. 543--558, 2021

    2021

  9. [17]

    R. Jena, D. Sethi, P. Chaudhari, and J. Gee, ``Deep learning in medical image registration: Magic or mirage?'' in The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024. [Online]. Available: https://openreview.net/forum?id=lZJ0WYI5YC

    2024

  10. [18]

    B. B. Avants, C. L. Epstein, M. Grossman, and J. C. Gee, `` Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain ,'' Medical Image Analysis, vol. 12, no. 1, pp. 26--41, 2008

    2008

  11. [19]

    Balakrishnan, A

    G. Balakrishnan, A. Zhao, M. R. Sabuncu, J. Guttag, and A. V. Dalca, ``Voxelmorph: a learning framework for deformable medical image registration,'' IEEE Trans. Med. Imag., vol. 38, no. 8, pp. 1788--1800, 2019

    2019

  12. [20]

    Torralba and A

    A. Torralba and A. A. Efros, ``Unbiased look at dataset bias,'' in 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011, pp. 1521--1528

    2011

  13. [21]

    K. Zhou, Z. Liu, Y. Qiao, T. Xiang, and C. C. Loy, ``Domain generalization: A survey,'' IEEE transactions on pattern analysis and machine intelligence, vol. 45, no. 4, pp. 4396--4415, 2022

    2022

  14. [22]

    Wang and W

    M. Wang and W. Deng, ``Deep visual domain adaptation: A survey,'' Neurocomputing, vol. 312, pp. 135--153, 2018

    2018

  15. [23]

    W. D. Penny, K. J. Friston, J. T. Ashburner, S. J. Kiebel, and T. E. Nichols, Statistical parametric mapping: the analysis of functional brain images. 1em plus 0.5em minus 0.4em Elsevier, 2011

    2011

  16. [24]

    Ashburner and K

    J. Ashburner and K. J. Friston, ``Voxel-based morphometry—the methods,'' Neuroimage, vol. 11, no. 6, pp. 805--821, 2000

    2000

  17. [25]

    Fischl, `` FreeSurfer ,'' NeuroImage, vol

    B. Fischl, `` FreeSurfer ,'' NeuroImage, vol. 62, no. 2, pp. 774--781, 2012

    2012

  18. [26]

    Younes, F

    L. Younes, F. Arrate, and M. I. Miller, ``Evolutions equations in computational anatomy,'' NeuroImage, vol. 45, no. 1, pp. S40--S50, 2009

    2009

  19. [27]

    Gregor and Y

    K. Gregor and Y. LeCun, ``Learning fast approximations of sparse coding,'' in Proceedings of the 27th international conference on international conference on machine learning, 2010, pp. 399--406

    2010

  20. [28]

    L. Tian, H. Greer, R. Kwitt, F.-X. Vialard, R. S. J. Est\' e par, S. Bouix, R. Rushmore, and M. Niethammer, `` uniGradICON: A foundation model for medical image registration ,'' in 27 ^ th International Conference on Medical Image Computing and Computer Assisted Intervention (...

    2024

  21. [29]

    Y. Liu, J. Chen, S. Wei, A. Carass, and J. L. Prince, `` On Finite Difference Jacobian Computation in Deformable Image Registration ,'' International Journal of Computer Vision, pp. 1--11, 2024

    2024

  22. [30]

    Greer, L

    H. Greer, L. Tian, F.-X. Vialard, R. Kwitt, S. Bouix, R. San Jose Estepar, R. Rushmore, and M. Niethammer, ``Inverse consistency by construction for multistep deep registration,'' in International Conference on Medical Image Computing and Computer-Assisted Intervention. 1em pl...

    2023

  23. [31]

    L. Tian, H. Greer, F. X. Vialard, R. Kwitt, R. S. J. Est\' e par, R. Rushmore, N. Makris, S. Bouix, and M. Niethammer, ``Gradicon: Approximate diffeomorphisms via gradient inverse consistency,'' in 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 202...

    2023

  24. [32]

    Rohlfing, `` Image similarity and tissue overlaps as surrogates for image registration accuracy: Widely used but unreliable ,'' IEEE Trans

    T. Rohlfing, `` Image similarity and tissue overlaps as surrogates for image registration accuracy: Widely used but unreliable ,'' IEEE Trans. Med. Imag., vol. 31, no. 2, pp. 153--163, 2012

    2012

  25. [33]

    Di Martino, C.-G

    A. Di Martino, C.-G. Yan, Q. Li, E. Denio, F. X. Castellanos, K. Alaerts, J. S. Anderson, M. Assaf, S. Y. Bookheimer, M. Dapretto et al., `` The autism brain imaging data exchange: Towards a large-scale evaluation of the intrinsic brain architecture in autism ,'' Molecular Psy...

    2014

  26. [34]

    Di Martino, D

    A. Di Martino, D. O’connor, B. Chen, K. Alaerts, J. S. Anderson, M. Assaf, J. H. Balsters, L. Baxter, A. Beggiato, S. Bernaerts et al., `` Enhancing studies of the connectome in autism using the autism brain imaging data exchange II ,'' Scientific Data, vol. 4, no. 1, pp. 1--15, 2017

    2017

  27. [35]

    `` IXI Dataset 2013; Brain Development --- brain-development.org ,'' https://brain-development.org/ixi-dataset

    2013

  28. [36]

    X.-N. Zuo, J. S. Anderson, P. Bellec, R. M. Birn, B. B. Biswal, J. Blautzik, J. Breitner, R. L. Buckner, V. D. Calhoun, F. X. Castellanos et al., ``An open science resource for establishing reliability and reproducibility in functional connectomics,'' Scientific Data, vol. 1, ...

    2014

  29. [37]

    A. J. Holmes, M. O. Hollinshead, T. M. O'Keefe , V. I. Petrov, G. R. Fariello, L. L. Wald, B. Fischl, B. R. Rosen, R. W. M. J. L. Roffman, J. W. Smoller, and R. L. Buckner, `` Brain Genomics Superstruct Project initial data release with structural, functional, and behavioral m...

    2015

  30. [38]

    S. A. Nastase, Y.-F. Liu, H. Hillman, A. Zadbood, L. Hasenfratz, N. Keshavarzian, J. Chen, C. J. Honey, Y. Yeshurun, M. Regev, M. Nguyen, C. H. C. Chang, C. Baldassano, O. Lositsky, E. Simony, M. A. Chow, Y. C. Leong, P. P. Brooks, E. Micciche, G. Choe, A. Goldstein, T. Vander...

    2021

  31. [39]

    Babayan, M

    A. Babayan, M. Erbey, D. Kumral, J. D. Reinelt, A. M. Reiter, J. R \"o bbig, H. L. Schaare, M. Uhlig, A. Anwander, P.-L. Bazin et al., `` A mind-brain-body dataset of MRI, EEG, cognition, emotion, and peripheral physiology in young and old adults ,'' Scientific Data, vol. 6, n...

    2019

  32. [40]

    Sunavsky and J

    A. Sunavsky and J. Poppenk, ``Neuroimaging predictors of creativity in healthy adults,'' NeuroImage, vol. 206, p. 116292, 2020

    2020

  33. [41]

    Li and R

    P. Li and R. B. Clariana, `` Reading comprehension in L1 and L2: An integrative approach ,'' Journal of Neurolinguistics, vol. 50, pp. 94--105, 2019

    2019

  34. [42]

    D. J. Follmer, S.-Y. Fang, R. B. Clariana, B. J. Meyer, and P. Li, `` What predicts adult readers’ understanding of STEM texts? '' Reading and Writing, vol. 31, pp. 185--214, 2018

    2018

  35. [43]

    D. P. Orfanos, V. Michel, Y. Schwartz, P. Pinel, A. Moreno, D. Le Bihan, and V. Frouin, ``The brainomics/localizer database,'' NeuroImage, vol. 144, pp. 309--314, 2017

    2017

  36. [44]

    Dufumier, A

    B. Dufumier, A. Grigis, J. Victor, C. Ambroise, V. Frouin, and E. Duchesnay, `` OpenBHB: A large-scale multi-site brain MRI data-set for age prediction and debiasing ,'' NeuroImage, vol. 263, p. 119637, 2022

    2022

  37. [45]

    D. S. Marcus, T. H. Wang, J. Parker, J. G. Csernansky, J. C. Morris, and R. L. Buckner, `` Open Access Series of Imaging Studies (OASIS): Cross-Sectional MRI Data in Young, Middle Aged, Nondemented, and Demented Older Adults ,'' Jrnl. of Cognitive Neuroscience, vol. 19, no. 9,...

    2007

  38. [46]

    A. Taha, G. Gilmore, M. Abbass, J. Kai, T. Kuehn, J. Demarco, G. Gupta, C. Zajner, D. Cao, R. Chevalier, A. Ahmed, A. Hadi, B. G. Karat, O. W. Stanley, P. J. Park, K. M. Ferko, D. Hemachandra, R. Vassallo, M. Jach, A. Thurairajah, S. Wong, M. C. tenorio, F. Ogunsanya, , A. R. ...

    2023

  39. [47]

    J. C. Reinhold, B. E. Dewey, A. Carass, and J. L. Prince, `` Evaluating the impact of intensity normalization on MR image synthesis ,'' in Medical Imaging 2019: Image Processing, vol. 10949. 1em plus 0.5em minus 0.4em International Society for Optics and Photonics, 2019, p. 109493H

    2019

  40. [48]

    Y. Huo, Z. Xu, Y. Xiong, K. Aboud, P. Parvathaneni, S. Bao, C. Bermudez, S. M. Resnick, L. E. Cutting, and B. A. Landman, `` 3D whole brain segmentation using spatially localized atlas network tiles ,'' NeuroImage, vol. 194, pp. 105--119, 2019

    2019

  41. [49]

    A. Taha, G. Gilmore, M. Abbass, J. Kai, T. Kuehn, J. Demarco, G. Gupta, C. Zajner, D. Cao, R. Chevalier et al., ``Magnetic resonance imaging datasets with anatomical fiducials for quality control and registration,'' Scientific Data, vol. 10, no. 1, p. 449, 2023

    2023

  42. [50]

    M. N. Lytle, R. Hammer, and J. R. Booth, `` A neuroimaging dataset on working memory and reward processing in children with and without ADHD ,'' Data in Brief, vol. 31, p. 105801, 2020

    2020

  43. [51]

    M. N. Lytle, D. D. Burman, and J. R. Booth, `` A neuroimaging dataset on response inhibition and selective attention in adults and children with and without ADHD ,'' Data in Brief, vol. 37, p. 107158, 2021

    2021

  44. [52]

    A. C. Nugent, A. G. Thomas, M. Mahoney, A. Gibbons, J. T. Smith, A. J. Charles, J. S. Shaw, J. D. Stout, A. M. Namyst, A. Basavaraj et al., `` The NIMH intramural healthy volunteer dataset: A comprehensive MEG, MRI, and behavioral resource ,'' Scientific Data, vol. 9, no. 1, p...

    2022

  45. [53]

    Mahler, J

    L. Mahler, J. Steiglechner, B. Bender, T. Lindig, D. Ramadan, J. Bause, F. Birk, R. Heule, E. Charyasz, M. Erb et al., `` UltraCortex: Submillimeter ultra-high field 9.4 t brain mr image collection and manual cortical segmentations ,'' bioRxiv, pp. 2024--06, 2024

    2024

  46. [54]

    V. S. Fonov, A. C. Evans, R. C. McKinstry, C. R. Almli, and D. Collins, ``Unbiased nonlinear average age-appropriate brain templates from birth to adulthood,'' NeuroImage , vol. 47, p. S102, 2009

    2009

  47. [55]

    M. P. Milham, L. Ai, B. Koo, T. Xu, C. Amiez, F. Balezeau, M. G. Baxter, E. L. Blezer, T. Brochier, A. Chen et al., ``An open resource for non-human primate imaging,'' Neuron, vol. 100, no. 1, pp. 61--74, 2018

    2018

  48. [56]

    Mazziotta, A

    J. Mazziotta, A. Toga, A. Evans, P. Fox, J. Lancaster, K. Zilles, R. Woods, T. Paus, G. Simpson, B. Pike, C. Holmes, L. Collins, P. Thompson, D. MacDonald, M. Iacoboni, T. Schormann, K. Amunts, N. Palomero-Gallagher, S. Geyer, L. Parsons, K. Narr, N. Kabani, G. L. Goualher, D....

    2001

  49. [57]

    B. B. Avants, N. Tustison, and H. Johnson, `` Advanced normalization tools (ANTS) ,'' Insight j, vol. 2, no. 365, pp. 1--35, 2009

    2009

  50. [58]

    Zhong, X

    T. Zhong, X. Wu, S. Liang, Z. Ning, L. Wang, Y. Niu, S. Yang, Z. Kang, Q. Feng, G. Li et al., `` nBEST: Deep-learning-based non-human primates Brain Extraction and Segmentation Toolbox across ages, sites and species ,'' NeuroImage, p. 120652, 2024

    2024

  51. [59]

    L. R. Dice, `` Measures of the Amount of Ecologic Association Between Species ,'' Ecology, vol. 26, no. 3, pp. 297--302, 1945

    1945

  52. [60]

    Carass, S

    A. Carass, S. Roy, A. Gherman, J. C. Reinhold, A. Jesson, T. Arbel, O. Maier, H. Handels, M. Ghafoorian, B. Platel, A. Birenbaum, H. Greenspan, D. L. Pham, C. M. Crainiceanu, P. A. Calabresi, J. L. Prince, W. R. Gray Roncal , R. T. Shinohara, and I. Oguz, `` Evaluating White M...

    2020

  53. [61]

    Hausdorff, Grundz\" u ge der Mengenlehre (Basics of Set Theory)

    F. Hausdorff, Grundz\" u ge der Mengenlehre (Basics of Set Theory) . 1em plus 0.5em minus 0.4em Leipzig, Germany: Verlag von Veit & Comp., 1914

    1914

  54. [62]

    B. Jian, J. Pan, M. Ghahremani, D. Rueckert, C. Wachinger, and B. Wiestler, ``Mamba? catch the hype or rethink what really helps for image registration,'' in International Workshop on Biomedical Image Registration. 1em plus 0.5em minus 0.4em Springer, 2024, pp. 86--97

    2024

  55. [63]

    Zhang, X

    H. Zhang, X. Chen, R. Hu, D. Liu, G. Li, and R. Wang, `` MemWarp: Discontinuity-Preserving Cardiac Registration with Memorized Anatomical Filters ,'' in 27 ^ th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2024) , ser. Lecture ...

    2024

  56. [64]

    X. Chen, Y. Xia, N. Ravikumar, and A. F. Frangi, `` CAR-Net: unsupervised Co-attention guided registration network for joint registration and structure learning ,'' arXiv preprint arXiv:2106.06637, 2021

    2021 arXiv

  57. [65]

    X. Jia, J. Bartlett, T. Zhang, W. Lu, Z. Qiu, and J. Duan, `` U-net vs transformer: Is u-net outdated in medical image registration? '' in Machine Learning in Medical Imaging (MLMI 2022)25 ^ th International Conference on Medical Image Computing and Computer Assisted Intervent...

    2022

  58. [66]

    Wagner, M

    F. Wagner, M. Thies, F. Denzinger, M. Gu, M. Patwari, S. Ploner, N. Maul, L. Pfaff, Y. Huang, and A. Maier, `` Trainable joint bilateral filters for enhanced prediction stability in low-dose CT ,'' Scientific Reports, vol. 12, no. 1, p. 17540, 2022

    2022

  59. [67]

    Honkamaa and P

    J. Honkamaa and P. Marttinen, `` SITReg: Multi-resolution architecture for symmetric, inverse consistent, and topology preserving image registration ,'' Machine Learning for Biomedical Imaging (MELBA), vol. 2, no. November 2024 Issue, pp. 2148--2194, 2024

    2024

  60. [68]

    Estienne, M

    T. Estienne, M. Vakalopoulou, E. Battistella, T. Henry, M. Lerousseau, A. Leroy, N. Paragios, and E. Deutsch, `` MICS: Multi-steps, inverse consistency and symmetric deep learning registration network ,'' arXiv preprint arXiv:2111.12123, 2021

    2021 arXiv

  61. [69]

    D. Gu, X. Cao, S. Ma, L. Chen, G. Liu, D. Shen, and Z. Xue, `` Pair-wise and group-wise deformation consistency in deep registration network ,'' in 23 ^ rd International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2020) , ser. Lecture Notes...

    2020

  62. [70]

    E. J. Hu, Y. Shen, P. Wallis, Z. Allen-Zhu, Y. Li, S. Wang, L. Wang, and W. Chen, `` Lora: Low-rank adaptation of large language models ,'' arXiv preprint arXiv:2106.09685, 2021

    2021 arXiv

  63. [71]

    Z. Tan, L. Zhang, Y. Lv, Y. Ma, and H. Lu, `` GroupMorph: medical image registration via grouping network with contextual fusion ,'' IEEE Trans. Med. Imag., vol. 43, no. 11, pp. 3807--3819, 2024

    2024

  64. [72]

    Balakrishnan, A

    G. Balakrishnan, A. Zhao, M. R. Sabuncu, J. Guttag, and A. V. Dalca, `` VoxelMorph: A learning framework for deformable medical image registration ,'' IEEE Trans. Med. Imag., vol. 38, no. 8, pp. 1788--1800, 2019

    2019

  65. [73]

    J. Chen, E. C. Frey, Y. He, W. P. Segars, Y. Li, and Y. Du, `` Transmorph: Transformer for unsupervised medical image registration ,'' Medical Image Analysis, vol. 82, p. 102615, 2022

    2022

  66. [74]

    Hwang, `` FAdam: Adam is a natural gradient optimizer using diagonal empirical Fisher information ,'' arXiv preprint arXiv:2405.12807, 2024

    D. Hwang, `` FAdam: Adam is a natural gradient optimizer using diagonal empirical Fisher information ,'' arXiv preprint arXiv:2405.12807, 2024

    2024 arXiv

  67. [75]

    Hiasa, Y

    Y. Hiasa, Y. Otake, M. Takao, T. Matsuoka, K. Takashima, A. Carass, J. L. Prince, N. Sugano, and Y. Sato, `` Cross-modality image synthesis from unpaired data using CycleGAN ,'' in Workshop on Simulation and Synthesis in Medical Imaging (SASHIMI) held in conjunction with the 2...

    2018

  68. [76]

    Greer, L

    H. Greer, L. Tian, F. Vialard, R. Kwitt, S. Bouix, R. San Jose Estepar , R. Rushmore, and M. Niethammer, `` Inverse consistency by construction for multistep deep registration ,'' in 26 ^ th International Conference on Medical Image Computing and Computer Assisted Intervention...

    2023

  69. [77]

    T. W. Mok and A. S. Chung, `` Fast symmetric diffeomorphic image registration with convolutional neural networks ,'' in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 4643--4652

    2020

  70. [78]

    H. J. Johnson and G. E. Christensen, `` Landmark and intensity-based, consistent thin-plate spline image registration ,'' in 17 ^ th Inf. Proc. in Med. Imaging (IPMI 2001) , ser. Lecture Notes in Computer Science, vol. 2082, 2001, pp. 329--343

    2001

  71. [79]

    Thirion, `` Image matching as a diffusion process: An analogy with Maxwell's demons ,'' Medical Image Analysis, vol

    J.-P. Thirion, `` Image matching as a diffusion process: An analogy with Maxwell's demons ,'' Medical Image Analysis, vol. 2, no. 3, pp. 243--260, 1998

    1998

  72. [80]

    Zhang, Y

    Y. Zhang, Y. Zhao, and Q. Tao, `` Improving Instance Optimization in Deformable Image Registration with Gradient Projection ,'' arXiv preprint arXiv:2410.15767, 2024

    2024 arXiv

  73. [81]

    T. Yu, S. Kumar, A. Gupta, S. Levine, K. Hausman, and C. Finn, `` Gradient surgery for multi-task learning ,'' Advances in Neural Information Processing Systems, vol. 33, pp. 5824--5836, 2020

    2020

  74. [82]

    H. Wang, Z. Wang, D. Ni, and Y. Wang, `` ModeTv2: GPU-accelerated Motion Decomposition Transformer for Pairwise Optimization in Medical Image Registration ,'' arXiv preprint arXiv:2403.16526, 2024

    2024

  75. [83]

    Avants, J

    B. Avants, J. T. Duda, H. Zhang, and J. C. Gee, `` Multivariate Normalization with Symmetric Diffeomorphisms for Multivariate Studies ,'' in 10 ^ th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2007) , ser. Lecture Notes in Com...

    2007

  76. [84]

    Avants and N

    B. Avants and N. Tustison, `` ANTs/ANTsR Brain Templates. figshare. Dataset. '' 2018. [Online]. Available: https://doi.org/10.6084/m9.figshare.915436.v2

    2018 doi

  77. [85]

    Klein, J

    A. Klein, J. Andersson, B. A. Ardekani, J. Ashburner, B. Avants, M. C. Chiang, G. E. Christensen, D. L. Collins, J. Gee, P. Hellier, J. H. Song, M. Jenkinson, C. Lepage, D. Rueckert, P. Thompson, T. Vercauteren, R. P. Woods, J. J. Mann, and R. V. Parsey, `` Evaluation of 14 no...

    2009

  78. [86]

    N. J. Tustison, B. B. Avants, L. Flors, T. A. Altes, E. E. de Lange , J. P. Mugler III , and J. C. Gee, `` Ventilation-based segmentation of the lungs using hyperpolarized ^ 3 He MRI ,'' Jrnl. of Magnetic Resonance Imaging, vol. 34, no. 4, pp. 831--841, 2011

    2011

  79. [87]

    M. P. Heinrich, M. Jenkinson, M. Brady, and J. A. Schnabel, `` MRF-based deformable registration and ventilation estimation of lung CT ,'' IEEE Trans. Med. Imag., vol. 32, no. 7, pp. 1239--1248, 2013

    2013

  80. [88]

    R. Jena, P. Chaudhari, and J. C. Gee, `` FireANTs: Adaptive Riemannian Optimization for Multi-Scale Diffeomorphic Registration ,'' arXiv preprint arXiv:2404.01249, 2024

    2024

  81. [89]

    Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, and B. Guo, ``Swin transformer: Hierarchical vision transformer using shifted windows,'' in Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 10\,012--10\,022

    2021

  82. [90]

    J. Chen, E. C. Frey, and Y. Du, ``Unsupervised learning of diffeomorphic image registration via transmorph,'' in International Workshop on Biomedical Image Registration. 1em plus 0.5em minus 0.4em Springer, 2022, pp. 96--102

    2022

  83. [91]

    Y. Liu, J. Chen, L. Zuo, A. Carass, and J. L. Prince, ``Vector field attention for deformable image registration,'' Journal of Medical Imaging, vol. 11, no. 6, pp. 064\,001--064\,001, 2024

    2024

  84. [92]

    Y. Liu, L. Zuo, S. Han, Y. Xue, J. L. Prince, and A. Carass, `` Coordinate Translator for Learning Deformable Medical Image Registration ,'' in Multiscale Multimodal Medical Imaging (MMMI 2022) held in conjunction with the 25 ^ th International Conference on Medical Image Comp...

    2022

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.