pith. sign in

arxiv: 2606.13315 · v1 · pith:RBJVJJYSnew · submitted 2026-06-11 · 💻 cs.CV · eess.IV

Masked and Predictive Self-Supervised Foundation Models for 3D Brain MRI

Esra Erg\"un , Hersh Chandarana , Dan Sodickson , G\"ozde \"Unal This is my paper

Pith reviewed 2026-06-27 07:31 UTC · model grok-4.3

classification 💻 cs.CV eess.IV
keywords self-supervised learningmasked autoencodersbrain MRIdisease detectionspectral regularizationJEPAfoundation models
0
0 comments X

The pith

Masked autoencoders with spectral-domain supervision outperform predictive models on brain MRI disease detection tasks.

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

The paper tests reconstruction-based masked autoencoders against predictive joint embedding architectures for self-supervised pretraining on heterogeneous 3D brain MRI scans. It introduces a spectral-domain reconstruction loss to the masked autoencoder to heighten sensitivity to fine anatomical detail and adds variance-covariance regularization to the predictive model to promote decorrelated features. Pretraining occurs in a contrast-agnostic manner on single-contrast volumes without modality concatenation. Evaluation across five downstream disease detection tasks shows that masked autoencoders equipped with the spectral loss deliver the strongest results overall. The findings indicate that each self-supervised objective carries a structural bias whose utility depends on how closely that bias matches the frequency and correlation properties of the target diagnostic signal.

Core claim

Masked autoencoders augmented with spectral-domain reconstruction loss achieve the highest performance across five MRI-based disease detection tasks, because the auxiliary spectral objective aligns directly with high-frequency anatomical structures that frequently encode the discriminative information, whereas standard reconstruction and predictive objectives with covariance regularization show more variable gains tied to the specific structure of each downstream task.

What carries the argument

Masked Autoencoder with spectral-domain reconstruction loss, which supplements standard masked image reconstruction by penalizing errors in the frequency domain to improve capture of fine-grained anatomical detail.

Load-bearing premise

The benefit of each self-supervised objective is determined by its relevance to the structure of the downstream task.

What would settle it

A disease detection task whose discriminative signal resides primarily in low-frequency components where the spectral-supervised MAE shows no consistent advantage over baseline MAE or JEPA variants.

Figures

Figures reproduced from arXiv: 2606.13315 by Dan Sodickson, Esra Erg\"un, G\"ozde \"Unal, Hersh Chandarana.

Figure 1
Figure 1. Figure 1: Overview of the proposed MAE pretraining and downstream evaluation framework. [PITH_FULL_IMAGE:figures/full_fig_p015_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of our JEPA framework. The student encoder processes visible tokens [PITH_FULL_IMAGE:figures/full_fig_p020_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Mean high-frequency power spectral density (PSD) comparison between UCSF [PITH_FULL_IMAGE:figures/full_fig_p025_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: t-SNE visualizations of encoder embeddings on the held-out test set ( [PITH_FULL_IMAGE:figures/full_fig_p036_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: k-shot classification performance (AUC) across methods on ADNI. Results are reported on a random subset of the test fold, which is shared on our github repository. On both NC vs. AD and NC vs. MCI tasks, SL-MAE outperforms ResNet across all k values, and VCR-JEPA outperforms ResNet except when k = 32 [PITH_FULL_IMAGE:figures/full_fig_p037_5.png] view at source ↗
read the original abstract

Self-supervised foundation models have shown strong promise in medical imaging. However, existing MRI foundation-model studies have primarily emphasized segmentation and dense prediction tasks, while systematic investigation of self-supervised foundation models for MRI-based disease detection remains limited. In this work, we investigate two major self-supervised pretraining paradigms for MRI-based disease detection: reconstruction-based learning via Masked Autoencoders (MAE) and predictive representation learning via Joint Embedding Predictive Architectures (JEPA). We study the role of auxiliary objectives by introducing a novel spectral-domain reconstruction loss for MAE to enhance sensitivity to fine-grained anatomical structure, and by integrating variance--covariance regularization (VCR) within our JEPA framework to encourage decorrelated latent representations. Our models are pretrained on heterogeneous single-contrast MRI volumes in a contrast-agnostic setting, without modality concatenation. Across five downstream disease detection tasks, our results highlight the importance of self-supervised objective design for medical foundation model pretraining, demonstrating that the downstream benefit of each objective is determined by its relevance to the task's structure. Specifically, spectral regularization yields the largest improvements when the downstream discriminative signal is characterized by strong high-frequency anatomical structures, while covariance regularization is most beneficial when discriminative information spans multiple decorrelated feature dimensions. MAE with spectral-domain supervision consistently achieves superior downstream performance for MRI-based disease detection. These findings suggest that self-supervised objectives in medical imaging encode specific biases, and their downstream benefit is fundamentally conditioned on the task's structure.

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 manuscript investigates two self-supervised pretraining paradigms for 3D brain MRI foundation models—reconstruction via Masked Autoencoders (MAE) and predictive learning via Joint Embedding Predictive Architectures (JEPA)—pretrained contrast-agnostically on heterogeneous single-contrast volumes. It introduces a spectral-domain reconstruction loss for MAE and variance-covariance regularization (VCR) for JEPA, claiming that MAE with spectral supervision yields superior performance across five downstream disease detection tasks, with each objective's benefit conditioned on task structure (high-frequency anatomical detail for spectral loss; decorrelated features for VCR).

Significance. If the empirical results are shown to isolate the contribution of each objective under matched training conditions, the work would demonstrate that self-supervised objectives encode specific inductive biases relevant to medical imaging tasks. This could inform more targeted pretraining strategies for disease detection in MRI, moving beyond generic segmentation-focused models by linking objective design to high-frequency or multi-dimensional discriminative signals.

major comments (2)
  1. [Abstract] Abstract: the claim that 'MAE with spectral-domain supervision consistently achieves superior downstream performance for MRI-based disease detection' is load-bearing for the paper's central thesis yet is stated without any quantitative metrics, error bars, dataset sizes, or statistical tests, preventing assessment of effect size or reliability.
  2. [Experiments] Experiments (assumed §4–5): attribution of gains specifically to the spectral objective requires explicit confirmation that all variants (standard MAE, spectral-MAE, JEPA, JEPA+VCR) used identical architectures, pretraining compute budgets, optimizer settings, data augmentations, and hyperparameter search effort. Without this equivalence, performance differences cannot be isolated from uncontrolled variables.
minor comments (2)
  1. [Abstract] Abstract: the five downstream tasks are referenced but not named or briefly characterized (e.g., which rely on high-frequency vs. decorrelated signals), reducing clarity on how task structure determines objective benefit.
  2. [Methods] Notation: the spectral-domain loss and VCR terms should be defined with explicit equations early in the methods to allow readers to verify their implementation details.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments. We address each major point below and will revise the manuscript accordingly to improve clarity and experimental transparency.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that 'MAE with spectral-domain supervision consistently achieves superior downstream performance for MRI-based disease detection' is load-bearing for the paper's central thesis yet is stated without any quantitative metrics, error bars, dataset sizes, or statistical tests, preventing assessment of effect size or reliability.

    Authors: We agree that the abstract would benefit from quantitative support for the central claim. We will revise the abstract to include the mean AUC improvement (with standard deviation) across the five tasks, the total number of pretraining volumes, and a brief note on the statistical tests performed. This will allow readers to assess effect sizes directly while preserving conciseness. revision: yes

  2. Referee: [Experiments] Experiments (assumed §4–5): attribution of gains specifically to the spectral objective requires explicit confirmation that all variants (standard MAE, spectral-MAE, JEPA, JEPA+VCR) used identical architectures, pretraining compute budgets, optimizer settings, data augmentations, and hyperparameter search effort. Without this equivalence, performance differences cannot be isolated from uncontrolled variables.

    Authors: All variants were trained under strictly matched conditions: identical 3D ViT architecture, same pretraining dataset and volume count, identical compute budget (epochs, steps, and hardware), same optimizer (AdamW) and learning-rate schedule, identical augmentation pipeline, and equivalent hyperparameter search effort. We will add a dedicated subsection 'Matched Experimental Conditions' in §4 together with a table enumerating all shared settings to make this equivalence explicit and allow unambiguous attribution of gains to the objectives. revision: yes

Circularity Check

0 steps flagged

No circularity: purely empirical comparison with no derivations or fitted predictions

full rationale

The paper conducts an empirical study comparing MAE and JEPA pretraining variants on MRI data for downstream disease detection tasks. No mathematical derivations, equations, or parameter-fitting steps are described that could reduce to self-definition or fitted-input-as-prediction. Claims of superior performance rest on experimental results across five tasks rather than any chain that collapses by construction to inputs. Self-citations are not invoked as load-bearing uniqueness theorems or ansatzes. The work is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no details on any free parameters, axioms, or invented entities used in the models or losses.

pith-pipeline@v0.9.1-grok · 5804 in / 1265 out tokens · 31994 ms · 2026-06-27T07:31:44.651467+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

112 extracted references · 5 canonical work pages

  1. [1]

    2024 , howpublished=

    Video Representation Learning with Joint-Embedding Predictive Architectures , author=. 2024 , howpublished=

    2024

  2. [2]

    2024 , howpublished=

    Revisiting feature prediction for learning visual representations from video , author=. 2024 , howpublished=

    2024

  3. [3]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=

    Revisiting MAE Pre-training for 3D Medical Image Segmentation , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=. 2025 , doi=

    2025

  4. [4]

    2026 , howpublished =

    Self-Supervised Learning for Knee Osteoarthritis: Diagnostic Limitations and Prognostic Value of Hospital Data , author =. 2026 , howpublished =

    2026

  5. [5]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

    Learning Generalizable 3D Medical Image Representations from Mask-Guided Self-Supervision , author =. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , year =

  6. [6]

    2026 , howpublished =

    Radhachandran, Ashwath and Ivezi. 2026 , howpublished =

    2026

  7. [7]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=

    Masked Autoencoders Are Scalable Vision Learners , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=. 2022 , doi=

    2022

  8. [8]

    2024 , publisher=

    Cox, Joseph and Liu, Peng and Stolte, Skylar E and Yang, Yunchao and Liu, Kang and See, Kyle B and Ju, Huiwen and Fang, Ruogu , journal=. 2024 , publisher=

    2024

  9. [9]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=

    Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages=. 2023 , doi=

    2023

  10. [10]

    Asbjørn Munk and Jakob Ambsdorf and Sebastian Llambias and Mads Nielsen , booktitle=

  11. [11]

    Abstract:

    Ulrich, Constantin and Wald, Tassilo and Suprijadi, Jonathan and Ziegler, Sebastian and Nohel, Michal and Peretzke, Robin and K. Abstract:. Bildverarbeitung f. 2026 , publisher=

    2026

  12. [12]

    Revisiting

    Wald, Tassilo and Ulrich, Constantin and Lukyanenko, Stanislav and Goncharov, Andrei and Paderno, Alberto and Miller, Maximilian and Maerkisch, Leander and Jaeger, Paul and Maier-Hein, Klaus , booktitle=. Revisiting. 2025 , doi=

    2025

  13. [13]

    Multi-modal Vision Pre-training for Medical Image Analysis , year=

    Rui, Shaohao and Chen, Lingzhi and Tang, Zhenyu and Wang, Lilong and Liu, Mianxin and Zhang, Shaoting and Wang, Xiaosong , booktitle=. Multi-modal Vision Pre-training for Medical Image Analysis , year=

  14. [14]

    and Masurkar, Arjun V

    Zhu, Weicheng and Huang, Haoxu and Tang, Huanze and Musthyala, Rushabh and Yu, Boyang and Chen, Long and Vega, Emilio and O'Donnell, Thomas and Hayek, Reya and Kuohn, Lindsey and Dehkharghani, Seena and Frontera, Jennifer A. and Masurkar, Arjun V. and Melmed, Kara and Razavian, Narges , title =. Nat. Biomed. Eng. , year =. doi:10.1038/s41551-026-01668-w , issn =

    work page doi:10.1038/s41551-026-01668-w

  15. [15]

    and Aneja, Sanjay , TITLE =

    Avesta, Arman and Hossain, Sajid and Lin, MingDe and Aboian, Mariam and Krumholz, Harlan M. and Aneja, Sanjay , TITLE =. Bioengineering , VOLUME =. 2023 , NUMBER =

    2023

  16. [16]

    and Rusinek, Henry and Chen, Jingyun and Ben, Zhang and Zhu, Weicheng and Fernandez-Grande, Carlos and Razavian, Narges , TITLE =

    Liu, Sheng and Masurkar, Arjun V. and Rusinek, Henry and Chen, Jingyun and Ben, Zhang and Zhu, Weicheng and Fernandez-Grande, Carlos and Razavian, Narges , TITLE =. Sci. Rep. , VOLUME =. 2022 , NUMBER =. doi:10.1038/s41598-022-20674-x , ISSN =

    work page doi:10.1038/s41598-022-20674-x 2022

  17. [17]

    and Ashburner, John , title =

    Chalcroft, Liam and Crinion, Jenny and Price, Cathy J. and Ashburner, John , title =. Simulation and Synthesis in Medical Imaging (SASHIMI 2025) , series =. 2025 , publisher =

    2025

  18. [18]

    and Martel, Anne L

    Xu, Tony and Hosseini, Sepehr and Anderson, Chris and Rinaldi, Anthony and Krishnan, Rahul G. and Martel, Anne L. and Goubran, Maged , title =. npj Digit. Med. , volume =. 2025 , doi =

    2025

  19. [19]

    2025 , howpublished =

    Sim. 2025 , howpublished =

    2025

  20. [20]

    Medical Slice Transformer: Improved Diagnosis and Explainability on

    M. Medical Slice Transformer: Improved Diagnosis and Explainability on. 2024 , howpublished =

    2024

  21. [21]

    2025 , horpublished=

    Towards Fair Medical AI: Adversarial Debiasing of 3D CT Foundation Embeddings , author=. 2025 , horpublished=

    2025

  22. [22]

    Maxime Oquab and Timothée Darcet and Théo Moutakanni and Huy Vo and Marc Szafraniec and Vasil Khalidov and Pierre Fernandez and Daniel Haziza and Francisco Massa and Alaaeldin El-Nouby and Mahmoud Assran and Nicolas Ballas and Wojciech Galuba and Russell Howes and Po-Yao Huang and Shang-Wen Li and Ishan Misra and Michael Rabbat and Vasu Sharma and Gabriel...

  23. [23]

    2025 , volume=

    Ye, Yiwen and Zhang, Jianpeng and Chen, Ziyang and Xia, Yong , journal=. 2025 , volume=

    2025

  24. [24]

    Jiang, Jue and Rangnekar, Aneesh and Veeraraghavan, Harini , title =. Med. Phys. , volume =. doi:10.1002/mp.17541 , year =

    work page doi:10.1002/mp.17541

  25. [25]

    Image Anal

    Med. Image Anal. , volume =. 2024 , issn =. doi:10.1016/j.media.2023.103062 , author =

    work page doi:10.1016/j.media.2023.103062 2024

  26. [26]

    Advances in Neural Information Processing Systems , editor =

    Taleb, Aiham and Loetzsch, Winfried and Danz, Noel and Severin, Julius and Gaertner, Thomas and Bergner, Benjamin and Lippert, Christoph , title =. Advances in Neural Information Processing Systems , editor =

  27. [27]

    Alzheimer's Disease Neuroimaging Initiative (

    Petersen, R C and Aisen, P S and Beckett, L A and Donohue, M C and Gamst, A C and Harvey, D J and Jack, C R and Jagust, W J and Shaw, L M and Toga, A W and Trojanowski, J Q and Weiner, M W , journal=. Alzheimer's Disease Neuroimaging Initiative (. 2010 , publisher=. doi:10.1212/WNL.0b013e3181cb3e25 , pmid=

    work page doi:10.1212/wnl.0b013e3181cb3e25 2010

  28. [28]

    , author=

    The parkinson’s progression markers initiative (ppmi)–establishing a pd biomarker cohort. , author=. Ann. Clin. Transl. Neurol. , volume=. 2018 , doi =

    2018

  29. [29]

    de Verdier, Maria Correia and Saluja, Rachit and Gagnon, Louis and LaBella, Dominic and Baid, Ujjwall and Tahon, Nourel Hoda and Foltyn-Dumitru, Martha and Zhang, Jikai and Alafif, Maram and Baig, Saif and Chang, Ken and D'Anna, Gennaro and Deptula, Lisa and Gupta, Diviya and Haider, Muhammad Ammar and Hussain, Ali and Iv, Michael and Kontzialis, Marinos ...

    2024

  30. [30]

    and Benzinger, Tammie L.S

    LaMontagne, Pamela J. and Benzinger, Tammie L.S. and Morris, John C. and Keefe, Sarah and Hornbeck, Russ and Xiong, Chengjie and Grant, Elizabeth and Hassenstab, Jason and Moulder, Krista and Vlassenko, Andrei G. and Raichle, Marcus E. and Cruchaga, Carlos and Marcus, Daniel , title =. 2019 , howpublished =

    2019

  31. [31]

    Proceedings of the International Conference on Learning Representations (ICLR) , year =

    Time-to-Event Pretraining for 3D Medical Imaging , author=. Proceedings of the International Conference on Learning Representations (ICLR) , year =

  32. [32]

    Self-Supervised Alignment Learning For Medical Image Segmentation , year=

    Li, Haofeng and Ouyang, Yiming and Wan, Xiang , booktitle=. Self-Supervised Alignment Learning For Medical Image Segmentation , year=

  33. [33]

    and Landman, Bennett and Xu, Daguang and Nath, Vishwesh and Hatamizadeh, Ali , title=

    Tang, Yucheng and Yang, Dong and Li, Wenqi and Roth, Holger R. and Landman, Bennett and Xu, Daguang and Nath, Vishwesh and Hatamizadeh, Ali , title=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages =. 2022 , doi =

    2022

  34. [34]

    2023 , howpublished =

    Azad, Bobby and Azad, Reza and Eskandari, Sania and Bozorgpour, Afshin and Kazerouni, Amirhossein and Rekik, Islem and Merhof, Dorit , title =. 2023 , howpublished =

    2023

  35. [35]

    2024 , howpublished=

    A Survey on Trustworthiness in Foundation Models for Medical Image Analysis , author=. 2024 , howpublished=

    2024

  36. [36]

    Mazurowski , year=

    Haoyu Dong and Yuwen Chen and Hanxue Gu and Nicholas Konz and Yaqian Chen and Qihang Li and Maciej A. Mazurowski , year=

  37. [37]

    NeuroImage , volume =

    Ghamizi, Salah and Kanli, Georgia and Deng, Yu and Palissot, Valérie and Perquin, Magali and Keunen, Olivier , title =. NeuroImage , volume =. 2026 , doi =

    2026

  38. [38]

    , title =

    Zhang, Chaojie and Chen, Shengjia and Cigdem, Ozkan and Rajamohan, Haresh Rengaraj and Cho, Kyunghyun and Kijowski, Richard and Deniz, Cem M. , title =. Radiol. Artif. Intell. , year =

  39. [39]

    van der Maaten, Laurens and Hinton, Geoffrey , title =. J. Mach. Learn. Res. , year =

  40. [40]

    Medical Image Computing and Computer Assisted Intervention --

    Zhang, Shengjie and Chen, Xiang and Ren, Bohan and Yang, Haibo and Yu, Ziqi and Zhang, Xiao-Yong and Zhou, Yuan , title =. Medical Image Computing and Computer Assisted Intervention --. 2022 , publisher =

    2022

  41. [41]

    and Pattichis, Constantinos S

    Leandrou, Stephanos and Lamnisos, Demetris and Mamais, Ioannis and Kyriacou, Panicos A. and Pattichis, Constantinos S. , title =. Front. Aging Neurosci. , year =

  42. [42]

    Wang, Luoyu and Feng, Qi and Ge, Xiuhong and Chen, Fenyang and Yu, Bo and Chen, Bing and Liao, Zhengluan and Lin, Biying and Lv, Yating and Ding, Zhongxiang , title =. Front. Neurosci. , year =

  43. [43]

    Louis and Spreng, R

    Wearn, Alfie and Raket, Lars Lau and Collins, D. Louis and Spreng, R. Nathan , title =. Brain Commun. , year =

  44. [44]

    Lee, Hyunna and Lee, Seokjun and Kim, Kun Woo , title =. J. Psychiatry Neurosci. , year =

  45. [45]

    Multi-scale radiomic analysis of sub-cortical regions in

    Chaddad, Ahmad and Desrosiers, Christian and Toews, Matthew , journal=. Multi-scale radiomic analysis of sub-cortical regions in. 2017 , publisher=

    2017

  46. [46]

    International Conference on Learning Representations (

    Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution , author=. International Conference on Learning Representations (. 2022 , doi=

    2022

  47. [47]

    2025 , howpublished =

    Keeping Representation Similarity in Finetuning for Medical Image Analysis , author=. 2025 , howpublished =

    2025

  48. [48]

    International Conference on Learning Representations , year=

    Better Fine-Tuning by Reducing Representational Collapse , author=. International Conference on Learning Representations , year=

  49. [49]

    International Conference on Learning Representations , year =

    Bardes, Adrien and Ponce, Jean and LeCun, Yann , title =. International Conference on Learning Representations , year =

  50. [50]

    Proceedings of the 38th International Conference on Machine Learning , series =

    Zbontar, Jure and Jing, Li and Misra, Ishan and LeCun, Yann and Deny, Stephane , title =. Proceedings of the 38th International Conference on Machine Learning , series =. 2021 , pages =

    2021

  51. [51]

    and Fox, Nick C

    Frisoni, Giovanni B. and Fox, Nick C. and Jack, Clifford R. and Scheltens, Philip and Thompson, Paul M. , title =. Nat. Rev. Neurol. , volume =. 2010 , doi =

    2010

  52. [52]

    Structural progression of

    Planche, Vincent and Manjon, Jos\'. Structural progression of. Brain Commun. , volume =. 2022 , doi =

    2022

  53. [53]

    Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural

    Bron, Esther E and Smits, Marion and van der Flier, Wiesje M and Vrenken, Hugo and Barkhof, Frederik and Scheltens, Philip and Papma, Janne M and Steketee, Rebecca M E and M. Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural. NeuroImage , volume =. 2015 , doi =

    2015

  54. [54]

    and Hwang, Kristy S

    Apostolova, Liana G. and Hwang, Kristy S. and Kohannim, Omid and Avila, David and Elashoff, David and Jack, Clifford R. and Shaw, Leslie and Trojanowski, John Q. and Weiner, Michael W. and Thompson, Paul M. , title =. Neurology , volume =. 2015 , doi =

    2015

  55. [55]

    and Mills, Samantha J

    Rose, Chris J. and Mills, Samantha J. and O'Connor, James P. B. and Buonaccorsi, Giovanni A. and Roberts, Caleb and Watson, Yvonne and Cheung, Susan and Zhao, Sha and Whitcher, Brandon and Jackson, Alan and Parker, Geoffrey J. M. , title =. Magn. Reson. Med. , volume =. 2009 , doi =

    2009

  56. [56]

    and Kinahan, Paul E

    Gillies, Robert J. and Kinahan, Paul E. and Hricak, Hedvig , title =. Radiology , volume =. 2016 , doi =

    2016

  57. [57]

    and Hallikainen, Merja and Vanhanen, Matti and Nissinen, Aulikki and Helkala, Eeva-Liisa and Vainio, Pauli and Vanninen, Ritva and Partanen, Kaarina and Soininen, Hilkka , title =

    Pennanen, Corina and Kivipelto, Miia and Tuomainen, Sari and Hartikainen, Paivi and Hanninen, Tuomo and Laakso, Mikko P. and Hallikainen, Merja and Vanhanen, Matti and Nissinen, Aulikki and Helkala, Eeva-Liisa and Vainio, Pauli and Vanninen, Ritva and Partanen, Kaarina and Soininen, Hilkka , title =. Neurobiol. Aging , volume =. 2004 , doi =

    2004

  58. [58]

    Postema, Merel C. and van Rooij, Daan and Anagnostou, Evdokia and Arango, Celso and Auzias, Guillaume and Behrmann, Marlene and Filho, Geraldo Busatto and Calderoni, Sara and Calvo, Rosa and Daly, Eileen and Deruelle, Christine and Di Martino, Adriana and Dinstein, Ilan and Duran, Fabio Luis S. and Durston, Sarah and Ecker, Christine and Ehrlich, Stefan a...

    2019

  59. [59]

    and Smits, Marion and Papma, Janne M

    Bron, Esther E. and Smits, Marion and Papma, Janne M. and Steketee, Rebecca M. E. and Meijboom, Rozanna and de Groot, Marius and van Swieten, John C. and Niessen, Wiro J. and van der Lugt, Aad , title =. NeuroImage Clin. , volume =. 2016 , doi =

    2016

  60. [60]

    Proceedings of the 37th International Conference on Machine Learning , articleno =

    Chen, Ting and Kornblith, Simon and Norouzi, Mohammad and Hinton, Geoffrey , title =. Proceedings of the 37th International Conference on Machine Learning , articleno =. 2020 , publisher =

    2020

  61. [61]

    van Rooij, Daan and Anagnostou, Evdokia and Arango, Celso and Auzias, Guillaume and Behrmann, Marlene and Busatto, Geraldo F. and Calderoni, Sara and Daly, Eileen and Deruelle, Christine and Di Martino, Adriana and Dinstein, Ilan and Duran, Fabio Luis Souza and Durston, Sarah and Ecker, Christine and Fair, Damien and Fedor, Jennifer and Fitzgerald, Jackie...

    2018

  62. [62]

    Haar, Shlomi and Berman, Sigal and Behrmann, Marlene and Dinstein, Ilan , title =. Cereb. Cortex , volume =. 2016 , doi =

    2016

  63. [63]

    Feng, Feng and Wang, Pan and Zhao, Kun and Zhou, Bo and Yao, Hongxiang and Meng, Qingqing and Wang, Lei and Zhang, Zengqiang and Ding, Yanhui and Wang, Luning and An, Ningyu and Zhang, Xi and Liu, Yong , title =. Front. Aging Neurosci. , volume =. 2018 , doi =

    2018

  64. [64]

    PLoS ONE , volume =

    Kinoshita, Manabu and Sakai, Mio and Arita, Hideyuki and Shofuda, Tomoko and Chiba, Yasuyoshi and Kagawa, Naoki and Watanabe, Yoshiyuki and Hashimoto, Naoya and Fujimoto, Yasunori and Yoshimine, Toshiki and Nakanishi, Katsuyuki and Kanemura, Yonehiro , title =. PLoS ONE , volume =. 2016 , doi =

    2016

  65. [65]

    IEEE Trans

    Shao, Minye and Wang, Zeyu and Duan, Haoran and Huang, Yawen and Zhai, Bing and Wang, Shizheng and Long, Yang and Zheng, Yefeng , title =. IEEE Trans. Med. Imaging , year =

  66. [66]

    and Yan, C-G and Li, Q

    Di Martino, A. and Yan, C-G and Li, Q. and Denio, E. and Castellanos, F.X. and Alaerts, K. and Anderson, J.S. and Assaf, M. and Bookheimer, S.Y. and Dapretto, M. and Deen, B. and Delmonte, S. and Dinstein, I. and Ertl-Wagner, B. and Fair, D.A. and Gallagher, L. and Kennedy, D.P. and Keown, C.L. and Keysers, C. and Lainhart, J.E. and Lord, C. and Luna, B. ...

    2014

  67. [67]

    Differential diagnosis of mild cognitive impairment and

    S. Differential diagnosis of mild cognitive impairment and. NeuroImage Clin. , volume =. 2016 , doi =

    2016

  68. [68]

    and Alexander, Daniel C

    Mazher, Moona and Parker, Geoff J.M. and Alexander, Daniel C. , title =. npj Imaging , year =

  69. [69]

    Oquab, Maxime and Darcet, Timoth. Trans. Mach. Learn. Res. , year =

  70. [70]

    2007 , howpublished =

    2007

  71. [71]

    and Assaf, Michal and Balsters, Joshua H

    Di Martino, Adriana and O'Connor, David and Chen, Bosi and Alaerts, Kaat and Anderson, Jeffrey S. and Assaf, Michal and Balsters, Joshua H. and Baxter, Leslie and Beggiato, Anita and Bernaerts, Sylvie and Blanken, Laura M. E. and Bookheimer, Susan Y. and Braden, B. Blair and Byrge, Lisa and Castellanos, F. Xavier and Dapretto, Mirella and Delorme, Richard...

    2017

  72. [72]

    and Benzinger, Tammie L.S

    LaMontagne, Pamela J. and Benzinger, Tammie L.S. and Morris, John C. and Keefe, Sarah and Hornbeck, Russ and Xiong, Chengjie and Grant, Elizabeth and Hassenstab, Jason and Moulder, Krista and Vlassenko, Andrei and Raichle, Marcus E. and Cruchaga, Carlos and Marcus, Daniel , title =. 2019 , howpublished =

    2019

  73. [73]

    , title =

    de Verdier, Maria Correia and Saluja, Rachit and Gagnon, Louis and LaBella, Dominic and Baid, Ujjwal et al. , title =. 2024 , howpublished =

    2024

  74. [74]

    Zimmerer, David and Full, Peter M. and Isensee, Fabian and Jäger, Paul and Adler, Tim and Petersen, Jens and Köhler, Gregor and Ross, Tobias and Reinke, Annika and Kascenas, Antanas and Jensen, Bjørn Sand and O’Neil, Alison Q. and Tan, Jeremy and Hou, Benjamin and Batten, James and Qiu, Huaqi and Kainz, Bernhard and Shvetsova, Nina and Fedulova, Irina and...

    2022

  75. [75]

    The University of California San Francisco Preoperative Diffuse Glioma

    Calabrese, Evan and Villanueva-Meyer, Javier E and Rudie, Jeffrey D and Rauschecker, Andreas M and Baid, Ujjwal and Bakas, Spyridon and Cha, Soonmee and Mongan, John T and Hess, Christopher P , journal=. The University of California San Francisco Preoperative Diffuse Glioma. 2022 , publisher=

    2022

  76. [76]

    2020 , howpublished =

    2020

  77. [77]

    Fast robust automated brain extraction , author=. Hum. Brain Mapp. , volume=. 2002 , publisher=

    2002

  78. [78]

    2020 , version =

    Brett, Matthew and Markiewicz, Christopher J and Hanke, Michael and C. 2020 , version =

    2020

  79. [79]

    Advances in Neural Information Processing Systems (NeurIPS) , volume =

    Dong, Zijian and Li, Ruilin and Wu, Yilei and Nguyen, Thuan Tinh and Chong, Joanna Su Xian and Ji, Fang and Tong, Nathanael Ren Jie and Chen, Christopher Li Hsian and Zhou, Juan Helen , title =. Advances in Neural Information Processing Systems (NeurIPS) , volume =. 2024 , doi =

    2024

  80. [80]

    Leslie Lamport, : a document preparation system, Addison Wesley, Massachusetts, 2nd edition, 1994

    1994

Showing first 80 references.