arXiv Computer Vision

A generalizable structural brain MRI foundation model built through dual-priority federated pretraining

BrainFedFM is a structural brain MRI foundation model that was federatively pretrained on 164,707 3‑D scans from 42 sites using a dual‑priority approach that emphasizes informative anatomical regions locally and prioritizes site contributions globally. The model outperformed seven baseline models—including four centralized foundation models—across 20 downstream tasks (classification, regression, segmentation), achieving a mean rank of 1.68 and a 50% performance gain, especially in classification and regression and among underrepresented populations. These results demonstrate the model’s generalizability and show that federated pretraining can effectively develop neuroimaging foundation models without pooling raw images.

arXiv AI
Sep 28

Combining General and Domain-Specific Pretext Tasks for Brain MR Image Segmentation

The paper investigates combining a domain‑specific self‑supervised task—voxel‑level brain age prediction—with a general task—image inpainting—to pretrain models for brain MRI segmentation. A multitask pretraining framework jointly optimizes both objectives, yielding representations that outperform single‑task pretraining and training from scratch on three segmentation benchmarks (multiple sclerosis lesions, ischemic stroke lesions, and cortical structures). The study demonstrates that integrating domain‑specific and general self‑supervised tasks benefits the development of generalizable neuroimaging foundation models.

By Tasneem Nasser, Susanne Schmid, Roberto Souza, Naser El-Sheimy
arXiv Machine Learning
Sep 24

A Scaling Study for fMRI Foundation Models

The study investigates how data volume, model size, and training duration affect the performance of fMRI foundation models. Using over 200 datasets and 10,000 GPU‑hours, the authors find that larger models benefit more from additional data, and that at a fixed compute budget, increasing data yields greater gains than enlarging the model. By selecting optimal combinations of data, size, and duration, they produce models that outperform existing fMRI foundation models on out‑of‑distribution tasks while requiring less pretraining compute.

By Wenhao Ye, Xuanye Pan, Junfeng Xia, Junxiang Zhang, Mo Wang, Quanying Liu
arXiv AI
Jun 12

BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning

arXiv:2604. 27277v3 Announce Type: replace-cross Abstract: Brain MRI underpins a wide range of neuroscientific and clinical applications, yet most learning-based methods remain task-specific and require substantial labeled data.

By Yizhou Wu, Shansong Wang, Yuheng Li, Mojtaba Safari, Mingzhe Hu, Chih-Wei Chang, Harini Veeraraghavan, Xiaofeng Yang
arXiv Computer Vision
Sep 14

Learning Sparse Latent Predictive Foundation Model for Multimodal Neuroimaging

The paper introduces Neuro‑JEPA, a sparse multimodal foundation model that learns unified representations of brain MRI across T1w, T2w, and FLAIR sequences using a latent predictive objective and a Mixture‑of‑Experts architecture. It was pretrained on over 1.5 million scans from 428,647 studies and systematically evaluates architectural, masking, objective, and sparsity choices for robust multimodal representation learning. Across 47 tasks from three health systems and 12 public datasets, Neuro‑JEPA consistently outperforms a simple CNN baseline, demonstrating its effectiveness for both clinical and research applications.

By Haoxu Huang, Long Chen, Jingyun Chen, Jinu Hyun, James Ryan Loftus, Kara Melmed, Daniel Orringer, Jennifer Frontera, Seena Dehkharghani, Arjun Masurkar, Narges Razavian
arXiv Computer Vision
Oct 28

Systematic Morphometric Brain Autoencoding for Representation Learning in Neurodegenerative Diseases

The paper introduces SYMBA, a self‑supervised pretraining framework tailored for structural T1‑weighted brain MRI. SYMBA uses anatomically constrained masking, a multi‑scale structural similarity reconstruction loss, and a progressive curriculum to capture distributed morphometric patterns. Evaluated on nine heterogeneous cohorts and two unseen datasets, SYMBA outperforms existing neuroimaging foundation models, achieving high AUROC scores for Alzheimer’s disease classification, mild cognitive impairment conversion, and amyloid status prediction, and shows strong sample efficiency in low‑label settings.

By Nuria Balb\'as-Esteban, Albert Belenguer-Llorens, Elaheh Moradi, Jussi Tohka, Vanessa G\'omez-Verdejo, Carlos Sevilla-Salcedo
arXiv AI
Nov 6

BrainPFN: Amortized Brain-Behaviour Prediction via Prior-Data Fitted Networks

BrainPFN is a Prior Data Fitted Network that predicts behavioral phenotypes from functional MRI data without needing per-score or per-cohort refitting. It is meta-trained on a domain-specific prior derived from about 7,000 real functional connectivity matrices, generating one billion synthetic training samples. In tests across four independent cohorts, BrainPFN outperforms Kernel Ridge Regression on small-sample predictions and rivals larger foundation models, all while using a modest, publicly available training pool.

By Carlo Alberto Barbano (MIND), Demian Wassermann (MIND)