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 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 Computer Vision
Oct 16

FlatClip: Reusing Image Foundation Models for fMRI Representation Learning via Cortical Flatmaps

FlatClip is a method that leverages frozen image foundation models to predict fMRI activity by mapping cortical signals onto anatomy‑linked flatmaps and extracting features with a SigLIP2 encoder. Only a lightweight downstream probe is trained, avoiding fMRI‑specific encoder pretraining or fine‑tuning. Experiments on four resting‑state benchmarks show FlatClip surpasses ROI‑level baselines on HCP and ADNI, while its performance varies on PPMI and remains below the best voxel‑level models; on NSD visual‑fMRI decoding, restricting inputs to task‑active cortex improves recognition, and spatial perturbations confirm the importance of cortical arrangement and input‑region selection.

By Mo Wang, Wenhao Ye, Zihan Ning, Jiayu Zuo, Junfeng Xia, Hongkai Wen, Quanying Liu