SLIM-Brain: A Data- and Training-Efficient Foundation Model for fMRI Data Analysis

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Mo, Xia, Junfeng, Ye, Wenhao, Liu, Enyu, Peng, Kaining, Feng, Jianfeng, Liu, Quanying, Wen, Hongkai
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917234207621120
author Wang, Mo
Xia, Junfeng
Ye, Wenhao
Liu, Enyu
Peng, Kaining
Feng, Jianfeng
Liu, Quanying
Wen, Hongkai
author_facet Wang, Mo
Xia, Junfeng
Ye, Wenhao
Liu, Enyu
Peng, Kaining
Feng, Jianfeng
Liu, Quanying
Wen, Hongkai
contents Foundation models are emerging as a powerful paradigm for fMRI analysis, but current approaches face a dual bottleneck of data- and training-efficiency. Atlas-based methods aggregate voxel signals into fixed regions of interest, reducing data dimensionality but discarding fine-grained spatial details, and requiring extremely large cohorts to train effectively as general-purpose foundation models. Atlas-free methods, on the other hand, operate directly on voxel-level information - preserving spatial fidelity but are prohibitively memory- and compute-intensive, making large-scale pre-training infeasible. We introduce SLIM-Brain (Sample-efficient, Low-memory fMRI Foundation Model for Human Brain), a new atlas-free foundation model that simultaneously improves both data- and training-efficiency. SLIM-Brain adopts a two-stage adaptive design: (i) a lightweight temporal extractor captures global context across full sequences and ranks data windows by saliency, and (ii) a 4D hierarchical encoder (Hiera-JEPA) learns fine-grained voxel-level representations only from the top-$k$ selected windows, while deleting about 70% masked patches. Extensive experiments across seven public benchmarks show that SLIM-Brain establishes new state-of-the-art performance on diverse tasks, while requiring only 4 thousand pre-training sessions and approximately 30% of GPU memory comparing to traditional voxel-level methods.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21881
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SLIM-Brain: A Data- and Training-Efficient Foundation Model for fMRI Data Analysis
Wang, Mo
Xia, Junfeng
Ye, Wenhao
Liu, Enyu
Peng, Kaining
Feng, Jianfeng
Liu, Quanying
Wen, Hongkai
Computer Vision and Pattern Recognition
Neurons and Cognition
Foundation models are emerging as a powerful paradigm for fMRI analysis, but current approaches face a dual bottleneck of data- and training-efficiency. Atlas-based methods aggregate voxel signals into fixed regions of interest, reducing data dimensionality but discarding fine-grained spatial details, and requiring extremely large cohorts to train effectively as general-purpose foundation models. Atlas-free methods, on the other hand, operate directly on voxel-level information - preserving spatial fidelity but are prohibitively memory- and compute-intensive, making large-scale pre-training infeasible. We introduce SLIM-Brain (Sample-efficient, Low-memory fMRI Foundation Model for Human Brain), a new atlas-free foundation model that simultaneously improves both data- and training-efficiency. SLIM-Brain adopts a two-stage adaptive design: (i) a lightweight temporal extractor captures global context across full sequences and ranks data windows by saliency, and (ii) a 4D hierarchical encoder (Hiera-JEPA) learns fine-grained voxel-level representations only from the top-$k$ selected windows, while deleting about 70% masked patches. Extensive experiments across seven public benchmarks show that SLIM-Brain establishes new state-of-the-art performance on diverse tasks, while requiring only 4 thousand pre-training sessions and approximately 30% of GPU memory comparing to traditional voxel-level methods.
title SLIM-Brain: A Data- and Training-Efficient Foundation Model for fMRI Data Analysis
topic Computer Vision and Pattern Recognition
Neurons and Cognition
url https://arxiv.org/abs/2512.21881