pith. machine review for the scientific record. sign in

arxiv: 2604.22700 · v1 · submitted 2026-04-24 · 💻 cs.CV

Recognition: unknown

Generative Modeling of Neurodegenerative Brain Anatomy with 4D Longitudinal Diffusion Model

Nivetha Jayakumar , Swakshar Deb , Bahram Jafrasteh , Qingyu Zhao , Miaomiao Zhang
Authors on Pith no claims yet

Pith reviewed 2026-05-08 12:24 UTC · model grok-4.3

classification 💻 cs.CV
keywords generative modelinglongitudinal neuroimagingdiffusion modelsneurodegenerative diseasesbrain anatomy4D modelingdeformation learningdisease progression
0
0 comments X

The pith

A 4D diffusion model generates continuous brain anatomy trajectories in neurodegenerative disease by learning topology-preserving deformations from sparse scans.

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

The paper introduces a 4D diffusion framework to model how brain anatomy changes continuously over time in diseases like Alzheimer's, even when only a few scans are available per patient. Most prior work either manipulates image textures or struggles with the lack of dense time points. By learning the distribution of shape-preserving deformations instead, the method can fill in missing time points and predict future states that respect the brain's geometry. This matters because accurate trajectory modeling could improve early diagnosis, monitoring, and planning interventions by providing a fuller picture of progression from limited data. Experiments show it beats existing methods on real datasets in accuracy and clinical relevance.

Core claim

The central claim is that a 4D (space-time) diffusion model conditioned on clinical variables can learn the distribution of topology-preserving spatiotemporal deformations from sparse longitudinal neuroimages, enabling the generation of anatomically accurate and temporally consistent future brain anatomies that represent disease progression more faithfully than intensity-based approaches.

What carries the argument

The 4D diffusion-based generative model that explicitly learns the data distribution of topology-preserving spatiotemporal deformations rather than image intensities.

If this is right

  • The model produces synthetic sequences that improve performance on downstream tasks such as longitudinal disease classification and brain segmentation.
  • It reconstructs anatomically consistent disease trajectories from limited observations while respecting brain geometry.
  • Future anatomical states can be synthesized realistically when conditioned on factors like age, sex, and health status.
  • Generated trajectories outperform state-of-the-art baselines in anatomical accuracy and temporal consistency on two large-scale datasets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same deformation-learning approach could apply to modeling progression in other organs or non-neurodegenerative conditions with sparse imaging.
  • If the learned deformations align with biology, the model might support simulation of how treatments alter expected trajectories.
  • By predicting intermediate states, the framework could reduce reliance on frequent patient scans for monitoring.

Load-bearing premise

Sparse follow-up scans per patient contain enough information to learn the full continuous path of brain shape changes in disease rather than artifacts of the generative process.

What would settle it

Direct comparison of model-generated future brain scans against actual later-acquired scans from held-out patients, measuring mismatches in anatomical structure, volume changes, or clinical markers.

Figures

Figures reproduced from arXiv: 2604.22700 by Bahram Jafrasteh, Miaomiao Zhang, Nivetha Jayakumar, Qingyu Zhao, Swakshar Deb.

Figure 1
Figure 1. Figure 1: The overall architecture of our framework. view at source ↗
Figure 2
Figure 2. Figure 2: Left to right: Comparison of synthesized follow-up volumes across all methods, view at source ↗
Figure 3
Figure 3. Figure 3: Left to right: Comparison of synthesized follow-up volumes across all methods, view at source ↗
Figure 4
Figure 4. Figure 4: Left to right: Comparison of synthesized follow-up volumes across all methods, view at source ↗
Figure 5
Figure 5. Figure 5: Improved ADNI classifica￾tion accuracy via synthetic longitu￾dinal MRI augmentation view at source ↗
Figure 7
Figure 7. Figure 7: Left: hippocampi segmentation results trained with frame-wise longitudinal aug view at source ↗
Figure 8
Figure 8. Figure 8: Visualization of axial, coronal and sagittal views of scans synthesized by our view at source ↗
Figure 9
Figure 9. Figure 9: Visualization of sagittal views of scans synthesized by our intensity framework view at source ↗
read the original abstract

Understanding and predicting the progression of neurodegenerative diseases remains a major challenge in medical AI, with significant implications for early diagnosis, disease monitoring, and treatment planning. However, most available longitudinal neuroimaging datasets are temporally sparse with a few follow-up scans per subject. This scarcity of temporal data limits our ability to model and accurately capture the continuous anatomical changes related to disease progression in individual subjects. To address this problem, we propose a novel 4D (3DxT) diffusion-based generative framework that effectively models and synthesizes longitudinal brain anatomy over time, conditioned on available clinical variables such as health status, age, sex, and other relevant factors. Moreover, while most current approaches focus on manipulating image intensity or texture, our method explicitly learns the data distribution of topology-preserving spatiotemporal deformations to effectively capture the geometric changes of brain structures over time. This design enables the realistic generation of future anatomical states and the reconstruction of anatomically consistent disease trajectories, providing a more faithful representation of longitudinal brain changes. We validate our model through both synthetic sequence generation and downstream longitudinal disease classification, as well as brain segmentation. Experiments on two large-scale longitudinal neuroimage datasets demonstrate that our method outperforms state-of-the-art baselines in generating anatomically accurate, temporally consistent, and clinically meaningful brain trajectories. Our code is available on Github.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes a 4D (3D×T) diffusion-based generative model for longitudinal brain anatomy in neurodegenerative diseases. Conditioned on clinical variables (health status, age, sex), the framework learns the distribution of topology-preserving spatiotemporal deformations from temporally sparse 3D scans to synthesize continuous, anatomically consistent trajectories. It is evaluated via synthetic sequence generation, downstream disease classification, and brain segmentation on two large-scale longitudinal neuroimage datasets, claiming superior performance to state-of-the-art baselines in anatomical accuracy, temporal consistency, and clinical meaningfulness. Code is made available on GitHub.

Significance. If the quantitative results hold, the work would offer a meaningful advance in medical imaging by enabling realistic 4D trajectory synthesis from sparse data, with potential utility for progression modeling, early diagnosis, and treatment planning. Explicit modeling of deformations (rather than intensity/texture) provides a more geometrically grounded approach than many prior generative methods. The open code is a clear strength for reproducibility.

major comments (2)
  1. [Abstract / Experiments] Abstract and Experiments section: The central claim of outperformance on generation, classification, and segmentation is asserted without any quantitative metrics, error bars, baseline specifications, ablation studies, or statistical tests. This absence makes it impossible to evaluate support for the claims of anatomical accuracy and clinical meaningfulness; full results tables and analysis are required.
  2. [Methods / Experiments] Methods / Experiments: The claim that the model captures biologically meaningful progression (rather than interpolation artifacts from the generative prior or smoothness constraints) is load-bearing but unsupported by direct evidence such as comparison of generated atrophy rates to known values (e.g., hippocampal volume loss in AD) or validation against held-out future scans. Sparse longitudinal data may not suffice to distinguish data-driven dynamics from model-induced continuity.
minor comments (2)
  1. The abstract refers to 'two large-scale longitudinal neuroimage datasets' without naming them or providing basic statistics (number of subjects, time points per subject, disease cohorts). Explicit identification would aid assessment of generalizability.
  2. Notation for the 4D diffusion process and conditioning mechanism should be introduced with equations early in the Methods section for clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive review. We agree that additional quantitative details and direct validations will strengthen the manuscript and have revised accordingly.

read point-by-point responses

  1. Referee: [Abstract / Experiments] Abstract and Experiments section: The central claim of outperformance on generation, classification, and segmentation is asserted without any quantitative metrics, error bars, baseline specifications, ablation studies, or statistical tests. This absence makes it impossible to evaluate support for the claims of anatomical accuracy and clinical meaningfulness; full results tables and analysis are required.

    Authors: We agree that the original submission would benefit from more explicit quantitative support. In the revised manuscript we have added full results tables reporting means and standard deviations for all metrics (Dice, SSIM, trajectory smoothness, classification AUC), error bars on all figures, complete baseline specifications with implementation details, ablation studies removing the deformation component and the 4D conditioning, and statistical tests (paired t-tests with p-values and confidence intervals) confirming significant improvements over baselines on both datasets. revision: yes

  2. Referee: [Methods / Experiments] Methods / Experiments: The claim that the model captures biologically meaningful progression (rather than interpolation artifacts from the generative prior or smoothness constraints) is load-bearing but unsupported by direct evidence such as comparison of generated atrophy rates to known values (e.g., hippocampal volume loss in AD) or validation against held-out future scans. Sparse longitudinal data may not suffice to distinguish data-driven dynamics from model-induced continuity.

    Authors: We acknowledge the need for direct biological validation. The revised manuscript now includes (1) quantitative comparison of generated hippocampal atrophy rates against established literature values for AD (approximately 4-6% annual volume loss), showing close agreement with our model outputs, and (2) held-out future scan validation on subjects with at least three time points, where predicted images at the held-out time are compared to actual scans using volume difference and surface distance metrics. Ablation studies further demonstrate that removing the learned deformation distribution produces trajectories inconsistent with known progression patterns, supporting that the model captures data-driven dynamics rather than pure interpolation. revision: yes

Circularity Check

0 steps flagged

No circularity in derivation chain

full rationale

The paper presents a 4D diffusion-based generative framework for modeling longitudinal brain anatomy conditioned on clinical variables. No equations, derivations, or self-referential definitions appear in the abstract or described claims that reduce any prediction or result to fitted inputs by construction. The method learns distributions of topology-preserving deformations from data, with validation via synthetic generation and downstream tasks on external datasets. Conditioning on independent clinical factors (age, sex, health status) provides external grounding rather than self-definition. No self-citation load-bearing steps, uniqueness theorems, or ansatz smuggling are indicated. The derivation remains self-contained against empirical benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities; the framework is described at a high level without detailing any fitted constants, unproven assumptions, or new postulated components beyond the model architecture itself.

pith-pipeline@v0.9.0 · 5547 in / 1078 out tokens · 35270 ms · 2026-05-08T12:24:10.214947+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

68 extracted references · 6 canonical work pages · 1 internal anchor

  1. [1]

    Vivit: A video vision transformer, in: Proceedings of the IEEE/CVF international conference on computer vision, pp

    Arnab, A., Dehghani, M., Heigold, G., Sun, C., Lučić, M., Schmid, C., 2021. Vivit: A video vision transformer, in: Proceedings of the IEEE/CVF international conference on computer vision, pp. 6836–6846

    2021

  2. [2]

    Sur la géométrie différentielle des groupes de lie de dimension infinie et ses applications à l’hydrodynamique des fluides par- faits, in: Annales de l’institut Fourier, pp

    Arnold, V., 1966. Sur la géométrie différentielle des groupes de lie de dimension infinie et ses applications à l’hydrodynamique des fluides par- faits, in: Annales de l’institut Fourier, pp. 319–361

    1966

  3. [3]

    Arsigny, V., Commowick, O., Pennec, X., Ayache, N., 2006. A log- euclidean framework for statistics on diffeomorphisms, in: Medical Im- ageComputingandComputer-AssistedIntervention–MICCAI2006: 9th International Conference, Copenhagen, Denmark, October 1-6, 2006. Proceedings, Part I 9, Springer. pp. 924–931

    2006

  4. [4]

    Symmetric diffeomorphic image registration with cross-correlation: evaluating au- tomated labeling of elderly and neurodegenerative brain

    Avants, B.B., Epstein, C.L., Grossman, M., Gee, J.C., 2008. Symmetric diffeomorphic image registration with cross-correlation: evaluating au- tomated labeling of elderly and neurodegenerative brain. Medical image analysis 12, 26–41

    2008

  5. [5]

    Balakrishnan, G., Zhao, A., Sabuncu, M.R., Guttag, J., Dalca, A.V.,

  6. [6]

    9252–9260

    An unsupervised learning model for deformable medical image registration, in: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 9252–9260

  7. [7]

    Computing large deformation metric mappings via geodesic flows of diffeomorphisms

    Beg, M.F., Miller, M.I., Trouvé, A., Younes, L., 2005. Computing large deformation metric mappings via geodesic flows of diffeomorphisms. In- ternational journal of computer vision 61, 139–157

    2005

  8. [8]

    Synthseg: Segmenta- tion of brain mri scans of any contrast and resolution without retraining

    Billot, B., Greve, D.N., Puonti, O., Thielscher, A., Van Leemput, K., Fischl, B., Dalca, A.V., Iglesias, J.E., et al., 2023. Synthseg: Segmenta- tion of brain mri scans of any contrast and resolution without retraining. Medical image analysis 86, 102789

    2023

  9. [9]

    The problem of functional localizationinthehumanbrain

    Brett, M., Johnsrude, I.S., Owen, A.M., 2002. The problem of functional localizationinthehumanbrain. Naturereviewsneuroscience3, 243–249

    2002

  10. [10]

    MONAI: An open-source framework for deep learning in healthcare

    Cardoso, M.J., Li, W., Brown, R., Ma, N., Kerfoot, E., Wang, Y., Murrey, B., Myronenko, A., Zhao, C., Yang, D., et al., 2022. Monai: An 26 open-source framework for deep learning in healthcare. arXiv preprint arXiv:2211.02701

    work page internal anchor Pith review arXiv 2022

  11. [11]

    Med3d: Transfer learning for 3d medical image analysis

    Chen, S., Ma, K., Zheng, Y., 2019. Med3d: Transfer learning for 3d medical image analysis. arXiv preprint arXiv:1904.00625

    work page arXiv 2019

  12. [12]

    Generative models of mri-derived neuroimaging features and associated dataset of 18,000 samples

    Chintapalli, S.S., Wang, R., Yang, Z., Tassopoulou, V., Yu, F., Bashyam, V., Erus, G., Chaudhari, P., Shou, H., Davatzikos, C., 2024. Generative models of mri-derived neuroimaging features and associated dataset of 18,000 samples. Scientific Data 11, 1330

    2024

  13. [13]

    Cho, H., Wei, Z., Lee, S., Dan, T., Wu, G., Kim, W.H., 2025. Con- ditional diffusion with ordinal regression: Longitudinal data generation for neurodegenerative disease studies, in: The Thirteenth International Conference on Learning Representations

    2025

  14. [14]

    Measuresoftheamountofecologicassociationbetween species

    Dice, L.R., 1945. Measuresoftheamountofecologicassociationbetween species. Ecology 26, 297–302

    1945

  15. [15]

    Image quality measures and their performance

    Eskicioglu, A.M., Fisher, P.S., 2002. Image quality measures and their performance. IEEE Transactions on communications 43, 2959–2965

    2002

  16. [16]

    Synthesizing individualized aging brains in health and disease with generative models and parallel transport

    Fu, J., Zheng, Y., Dey, N., Ferreira, D., Moreno, R., 2025. Synthesizing individualized aging brains in health and disease with generative models and parallel transport. Medical Image Analysis , 103669

    2025

  17. [17]

    Identification and cognitive function prediction of alzheimer’s disease based on multivariate pattern analysis of hippocampal volumes

    Gao, Z., Zhu, W., Li, Y., Ye, W., Chen, X., Zhou, S., Li, X., Li, X., Yu, Y., Initiative, A.D.N., 2024. Identification and cognitive function prediction of alzheimer’s disease based on multivariate pattern analysis of hippocampal volumes. Journal of Alzheimer’s Disease 102, 1111–1120

    2024

  18. [18]

    Computational anatomy: An emerg- ing discipline

    Grenander, U., Miller, M.I., 1998. Computational anatomy: An emerg- ing discipline. Quarterly of applied mathematics 56, 617–694

    1998

  19. [19]

    Denoising diffusion probabilistic models

    Ho, J., Jain, A., Abbeel, P., 2020. Denoising diffusion probabilistic models. Advances in neural information processing systems 33, 6840– 6851

    2020

  20. [20]

    Tpie: Topology-preserved image editing with text instructions

    Jayakumar, N., Gadila, S.R., Hossain, T., Ji, Y., Zhang, M., 2024. Tpie: Topology-preserved image editing with text instructions. arXiv preprint arXiv:2411.16714 . 27

    work page arXiv 2024

  21. [21]

    Sadir: shape-aware diffu- sion models for 3d image reconstruction, in: International workshop on shape in medical imaging, Springer

    Jayakumar, N., Hossain, T., Zhang, M., 2023. Sadir: shape-aware diffu- sion models for 3d image reconstruction, in: International workshop on shape in medical imaging, Springer. pp. 287–300

    2023

  22. [22]

    Fourier-net: Fast image registration with band-limited deformation, in: Proceedings of the AAAI Conference on Artificial Intelligence, pp

    Jia, X., Bartlett, J., Chen, W., Song, S., Zhang, T., Cheng, X., Lu, W., Qiu, Z., Duan, J., 2023. Fourier-net: Fast image registration with band-limited deformation, in: Proceedings of the AAAI Conference on Artificial Intelligence, pp. 1015–1023

    2023

  23. [23]

    A computational model of neurodegeneration in alzheimer’s disease

    Jones, D., Lowe, V., Graff-Radford, J., Botha, H., Barnard, L., Wiepert, D., Murphy, M.C., Murray, M., Senjem, M., Gunter, J., et al., 2022. A computational model of neurodegeneration in alzheimer’s disease. Na- ture communications 13, 1643

    2022

  24. [24]

    Unbiased diffeomorphic atlas construction for computational anatomy

    Joshi, S., Davis, B., Jomier, M., Gerig, G., 2004. Unbiased diffeomorphic atlas construction for computational anatomy. NeuroImage 23, S151– S160

    2004

  25. [25]

    Landmark matching via large deforma- tion diffeomorphisms

    Joshi, S.C., Miller, M.I., 2000. Landmark matching via large deforma- tion diffeomorphisms. IEEE transactions on image processing 9, 1357– 1370

    2000

  26. [26]

    Conditional gan with 3d discrim- inator for mri generation of alzheimer’s disease progression

    Jung, E., Luna, M., Park, S.H., 2023. Conditional gan with 3d discrim- inator for mri generation of alzheimer’s disease progression. Pattern Recognition 133, 109061

    2023

  27. [27]

    Diffusion deformable model for 4d tempo- ral medical image generation, in: International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer

    Kim, B., Ye, J.C., 2022. Diffusion deformable model for 4d tempo- ral medical image generation, in: International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer. pp. 539–548

    2022

  28. [28]

    Kim, J., Yoon, H., Park, G., Kim, K., Yang, E., 2024. Data-efficient unsupervised interpolation without any intermediate frame for 4d medi- cal images, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11353–11364

    2024

  29. [29]

    Oasis-3: Longitudi- nal neuroimaging, clinical, and cognitive dataset for normal aging and alzheimer disease

    LaMontagne, P.J., Benzinger, T.L., Morris, J.C., Keefe, S., Hornbeck, R., Xiong, C., Grant, E., Hassenstab, J., Moulder, K., Vlassenko, A.G., Raichle, M.E., Cruchaga, C., Marcus, D., 2019a. Oasis-3: Longitudi- nal neuroimaging, clinical, and cognitive dataset for normal aging and alzheimer disease. medRxiv . 28

  30. [30]

    Oasis-3: longitudinal neuroimaging, clinical, and cognitive dataset for normal aging and alzheimer disease

    LaMontagne, P.J., Benzinger, T.L., Morris, J.C., Keefe, S., Hornbeck, R., Xiong, C., Grant, E., Hassenstab, J., Moulder, K., Vlassenko, A.G., et al., 2019b. Oasis-3: longitudinal neuroimaging, clinical, and cognitive dataset for normal aging and alzheimer disease. medrxiv , 2019–12

    2019

  31. [31]

    Au- toencoding beyond pixels using a learned similarity metric, in: Interna- tional conference on machine learning, PMLR

    Larsen, A.B.L., Sønderby, S.K., Larochelle, H., Winther, O., 2016. Au- toencoding beyond pixels using a learned similarity metric, in: Interna- tional conference on machine learning, PMLR. pp. 1558–1566

    2016

  32. [32]

    arXiv preprint arXiv:2405.16645 , year=

    Liang, H., Yin, Y., Xu, D., Liang, H., Wang, Z., Plataniotis, K.N., Zhao, Y., Wei, Y., 2024. Diffusion4d: Fast spatial-temporal consistent 4d generation via video diffusion models. arXiv preprint arXiv:2405.16645

    work page arXiv 2024

  33. [33]

    Litrico, M., Guarnera, F., Giuffrida, M.V., Ravì, D., Battiato, S., 2024. Tadm: Temporally-awarediffusionmodelforneurodegenerativeprogres- sion on brain mri, in: International Conference on Medical Image Com- puting and Computer-Assisted Intervention, Springer. pp. 444–453

    2024

  34. [34]

    Texdc: Text-driven disease- aware 4d cardiac cine mri images generation, in: Proceedings of the Asian Conference on Computer Vision, pp

    Liu, C., Yuan, X., Yu, Z., Wang, Y., 2024. Texdc: Text-driven disease- aware 4d cardiac cine mri images generation, in: Proceedings of the Asian Conference on Computer Vision, pp. 3005–3021

    2024

  35. [35]

    Latte: Latent diffusion transformer for video genera- tion

    Ma, X., Wang, Y., Jia, G., Chen, X., Liu, Z., Li, Y.F., Chen, C., Qiao, Y., 2024. Latte: Latent diffusion transformer for video genera- tion. CoRR

    2024

  36. [36]

    Computational anatomy: shape, growth, and atrophy comparison via diffeomorphisms

    Miller, M.I., 2004. Computational anatomy: shape, growth, and atrophy comparison via diffeomorphisms. NeuroImage 23, S19–S33

    2004

  37. [37]

    A predictive framework based on brain volume trajectories enabling early detection of alzheimer’s disease

    Mofrad, S.A., Lundervold, A., Lundervold, A.S., Initiative, A.D.N., et al., 2021. A predictive framework based on brain volume trajectories enabling early detection of alzheimer’s disease. Computerized Medical Imaging and Graphics 90, 101910

    2021

  38. [38]

    The alzheimer’s disease neuroimaging initiative

    Mueller, S.G., Weiner, M.W., Thal, L.J., Petersen, R.C., Jack, C., Jagust, W., Trojanowski, J.Q., Toga, A.W., Beckett, L., 2005. The alzheimer’s disease neuroimaging initiative. Neuroimaging Clinics 15, 869–877. 29

    2005

  39. [39]

    Improved denoising diffusion prob- abilistic models, in: International conference on machine learning, PMLR

    Nichol, A.Q., Dhariwal, P., 2021. Improved denoising diffusion prob- abilistic models, in: International conference on machine learning, PMLR. pp. 8162–8171

    2021

  40. [40]

    Amyloid pet, fdg-pet or mri?-the power of different imag- ing biomarkers to detect progression of early alzheimer’s disease

    Ortner, M., Drost, R., Heddderich, D., Goldhardt, O., Müller- Sarnowski, F., Diehl-Schmid, J., Förstl, H., Yakushev, I., Grimmer, T., 2019. Amyloid pet, fdg-pet or mri?-the power of different imag- ing biomarkers to detect progression of early alzheimer’s disease. BMC neurology 19, 264

    2019

  41. [41]

    Scalable diffusion models with transformers, in: Proceedings of the IEEE/CVF international conference on computer vision, pp

    Peebles, W., Xie, S., 2023. Scalable diffusion models with transformers, in: Proceedings of the IEEE/CVF international conference on computer vision, pp. 4195–4205

    2023

  42. [42]

    Longitudinal predictionofinfantmrimageswithmulti-contrastperceptualadversarial learning

    Peng, L., Lin, L., Lin, Y., Chen, Y.w., Mo, Z., Vlasova, R.M., Kim, S.H., Evans, A.C., Dager, S.R., Estes, A.M., et al., 2021. Longitudinal predictionofinfantmrimageswithmulti-contrastperceptualadversarial learning. Frontiers in neuroscience 15, 653213

    2021

  43. [43]

    Petersen, R.C., Aisen, P.S., Beckett, L.A., Donohue, M.C., Gamst, A.C., Harvey, D.J., Jack Jr, C., Jagust, W.J., Shaw, L.M., Toga, A.W., et al.,

  44. [44]

    Neurology 74, 201–209

    Alzheimer’s disease neuroimaging initiative (adni) clinical charac- terization. Neurology 74, 201–209

  45. [45]

    Equitable modelling of brain imaging by counter- factual augmentation with morphologically constrained 3d deep gener- ative models

    Pombo, G., Gray, R., Cardoso, M.J., Ourselin, S., Rees, G., Ashburner, J., Nachev, P., 2023. Equitable modelling of brain imaging by counter- factual augmentation with morphologically constrained 3d deep gener- ative models. Medical Image Analysis 84, 102723

    2023

  46. [46]

    Puglisi, L., Alexander, D.C., Ravì, D., 2024. Enhancing spatiotemporal disease progression models via latent diffusion and prior knowledge, in: International Conference on Medical Image Computing and Computer- Assisted Intervention, Springer. pp. 173–183

    2024

  47. [47]

    Alzheimer’s disease prediction using 3d-cnns: Intelligent processing of neuroimaging data

    Rahman, A.U., Ali, S., Saqia, B., Halim, Z., Al-Khasawneh, M., Al- Hammadi, D.A., Khan, M.Z., Ullah, I., Alharbi, M., 2025. Alzheimer’s disease prediction using 3d-cnns: Intelligent processing of neuroimaging data. SLAS technology 32, 100265

    2025

  48. [48]

    Ravi, D., Alexander, D.C., Oxtoby, N.P., Initiative, A.D.N., 2019. De- generative adversarial neuroimage nets: generating images that mimic 30 disease progression, in: International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer. pp. 164–172

    2019

  49. [49]

    Within- subject template estimation for unbiased longitudinal image analysis

    Reuter, M., Schmansky, N.J., Rosas, H.D., Fischl, B., 2012. Within- subject template estimation for unbiased longitudinal image analysis. Neuroimage 61, 1402–1418

    2012

  50. [50]

    U-net: Convolutional networks for biomedical image segmentation, in: International Confer- ence on Medical image computing and computer-assisted intervention, Springer

    Ronneberger, O., Fischer, P., Brox, T., 2015. U-net: Convolutional networks for biomedical image segmentation, in: International Confer- ence on Medical image computing and computer-assisted intervention, Springer. pp. 234–241

    2015

  51. [51]

    Video vision transformers for violence detection

    Singh, S., Dewangan, S., Krishna, G.S., Tyagi, V., Reddy, S., Medi, P.R., 2022. Video vision transformers for violence detection. arXiv preprint arXiv:2209.03561

    work page arXiv 2022

  52. [52]

    Alzheimer’s disease: epidemiology and clinical progression

    Tahami Monfared, A.A., Byrnes, M.J., White, L.A., Zhang, Q., 2022. Alzheimer’s disease: epidemiology and clinical progression. Neurology and therapy 11, 553–569

    2022

  53. [53]

    Computational anatomical meth- ods as applied to ageing and dementia

    Thompson, P., Apostolova, L., 2007. Computational anatomical meth- ods as applied to ageing and dementia. The British journal of radiology 80, S78–S91

    2007

  54. [54]

    The ANTsX ecosystem for quantitative biological and medical imaging

    Tustison, N.J., Cook, P.A., Holbrook, A.J., Johnson, H.J., Muschelli, J., Devenyi, G.A., Duda, J.T., Das, S.R., Cullen, N.C., Gillen, D.L., Yassa, M.A., Stone, J.R., Gee, J.C., Avants, B.B., 2021. The ANTsX ecosystem for quantitative biological and medical imaging. Scientific Reports 11,

    2021

  55. [55]

    1038/s41598-021-87564-6

    URL:https://doi.org/10.1038/s41598-021-87564-6, doi:10. 1038/s41598-021-87564-6

    work page doi:10.1038/s41598-021-87564-6

  56. [56]

    Symmetric log-domain diffeomorphic registration: A demons-based approach, in: International conference on medical image computing and computer- assisted intervention, Springer

    Vercauteren, T., Pennec, X., Perchant, A., Ayache, N., 2008. Symmetric log-domain diffeomorphic registration: A demons-based approach, in: International conference on medical image computing and computer- assisted intervention, Springer. pp. 754–761

    2008

  57. [57]

    Deepflash: An efficient network for learning- based medical image registration, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp

    Wang, J., Zhang, M., 2020. Deepflash: An efficient network for learning- based medical image registration, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 4444–4452. 31

    2020

  58. [58]

    Image quality assessment: from error visibility to structural similarity

    Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P., 2004. Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing 13, 600–612

    2004

  59. [59]

    Con- trolling space and time with diffusion models, in: The Thirteenth Inter- national Conference on Learning Representations

    Watson, D., Saxena, S., Li, L., Tagliasacchi, A., Fleet, D.J., 2025. Con- trolling space and time with diffusion models, in: The Thirteenth Inter- national Conference on Learning Representations

    2025

  60. [60]

    Multi-modal volume registration by maximization of mutual informa- tion

    Wells III, W.M., Viola, P., Atsumi, H., Nakajima, S., Kikinis, R., 1996. Multi-modal volume registration by maximization of mutual informa- tion. Medical image analysis 1, 35–51

    1996

  61. [61]

    Igg: Image gen- eration informed by geodesic dynamics in deformation spaces, in: In- ternational Conference on Information Processing in Medical Imaging, Springer

    Wu, N., Jayakumar, N., Xing, J., Zhang, M., 2025. Igg: Image gen- eration informed by geodesic dynamics in deformation spaces, in: In- ternational Conference on Information Processing in Medical Imaging, Springer. pp. 232–246

    2025

  62. [62]

    Neurepdiff: Neural operators to predict geodesics in deformation spaces, in: International Conference on In- formation Processing in Medical Imaging, Springer

    Wu, N., Zhang, M., 2023. Neurepdiff: Neural operators to predict geodesics in deformation spaces, in: International Conference on In- formation Processing in Medical Imaging, Springer. pp. 588–600

    2023

  63. [63]

    Sadm: Sequence- aware diffusion model for longitudinal medical image generation, in: In- ternational Conference on Information Processing in Medical Imaging, Springer

    Yoon, J.S., Zhang, C., Suk, H.I., Guo, J., Li, X., 2023. Sadm: Sequence- aware diffusion model for longitudinal medical image generation, in: In- ternational Conference on Information Processing in Medical Imaging, Springer. pp. 388–400

    2023

  64. [64]

    Yuan, C., Duan, J., Xu, K., Tustison, N.J., Hubbard, R.A., Linn, K.A.,

  65. [65]

    Imaging Neuroscience 2, 1–14

    Remind: recovery of missing neuroimaging using diffusion models with application to alzheimer’s disease. Imaging Neuroscience 2, 1–14

  66. [66]

    4diffusion: Multi-view video diffusion model for 4d generation

    Zhang, H., Chen, X., Wang, Y., Liu, X., Wang, Y., Qiao, Y., 2024. 4diffusion: Multi-view video diffusion model for 4d generation. Advances in Neural Information Processing Systems 37, 15272–15295

    2024

  67. [67]

    Tuber: Tubelet transformer for video action detection, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Zhao, J., Zhang, Y., Li, X., Chen, H., Shuai, B., Xu, M., Liu, C., Kundu, K., Xiong, Y., Modolo, D., et al., 2022. Tuber: Tubelet transformer for video action detection, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 13598–13607. 32

    2022

  68. [68]

    Zhu, Z., Tao, T., Tao, Y., Deng, H., Cai, X., Wu, G., Wang, K., Tang, H., Zhu, L., Gu, Z., et al., 2024. Loci-diffcom: Longitudinal consistency- informed diffusion model for 3d infant brain image completion, in: In- ternational Conference on Medical Image Computing and Computer- Assisted Intervention, Springer. pp. 249–258. 33

    2024