AREION: A Foundation Model for Atlas-Agnostic Representation Learning of Brain Connectome
- 1University of Pennsylvania
Abstract
Brain foundation models have shown promising advances in learning generalizable representations from functional magnetic resonance imaging (fMRI) data. However, existing models are typically tied to specific brain atlases and cannot be applied to fMRI data represented using different atlases without full retraining, thereby severely limiting their broader utility across neuroimaging studies. To address this bottleneck, we introduce AREION, atlas-agnostic representation learning of brain connectomes. AREION overcomes atlas dependency through two integrated innovations. First, a Brain Region Contextualizing technique maps brain regions and connectivity entries from diverse atlases into a unified representation space based on their spatial, functional, and morphological characteristics. Second, a Hierarchical Brain Transformer Encoder processes region-specific connectivity profiles as variable-length token sequences and jointly captures local functional structure and global brain functional organization, conditioned on brain regional context, without assuming a fixed number of regions. Together, these two techniques enable the learning of generalizable representations across brain atlases with different parcellation resolutions and connectivity dimensions. AREION is pretrained on a large-scale multi-atlas cohort of fMRI data via self-supervised modeling and evaluated across diverse downstream tasks. Extensive experiments show that it outperforms strong supervised baselines and fMRI foundation models while using substantially fewer parameters, demonstrating strong transferability to atlases unseen during pretraining, generalization to external datasets, and excellent interpretability at the brain regional level.
Related
- FlatClip: Reusing Image Foundation Models for fMRI Representation Learning via Cortical Flatmaps
- scRep: A Latent-Space Self-Distilled Foundation Model for Single-Cell Representation Learning
- FedFound: a federated foundation model for lifespan brain morphological connectome analysis
- OncoNoteBERT: A Foundation Representation Model for Natural Language Processing of Real-World Outpatient Oncology Notes
- MarkerScout: A Disease-Agnostic Machine Learning Framework for Biomarker Prediction from Multi-Scale Mechanistic Models
BibTeX
@misc{zhang2026areion,
title = {{AREION: A Foundation Model for Atlas-Agnostic Representation Learning of Brain Connectome}},
author = {Z. Zhang and H. Li and Y. Zhang and G. You and C. Wang and Y. Fan},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.26.754551},
url = {https://www.biorxiv.org/content/10.64898/2026.09.26.754551v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - AREION: A Foundation Model for Atlas-Agnostic Representation Learning of Brain Connectome
AU - Z. Zhang
AU - H. Li
AU - Y. Zhang
AU - G. You
AU - C. Wang
AU - Y. Fan
PY - 2026
DA - 2026/09/26
PB - bioRxiv
DO - 10.64898/2026.09.26.754551
UR - https://www.biorxiv.org/content/10.64898/2026.09.26.754551v1
AB - Brain foundation models have shown promising advances in learning generalizable representations from functional magnetic resonance imaging (fMRI) data. However, existing models are typically tied to specific brain atlases and cannot be applied to fMRI data represented using different atlases without full retraining, thereby severely limiting their broader utility across neuroimaging studies. To address this bottleneck, we introduce AREION, atlas-agnostic representation learning of brain connectomes. AREION overcomes atlas dependency through two integrated innovations. First, a Brain Region Contextualizing technique maps brain regions and connectivity entries from diverse atlases into a unified representation space based on their spatial, functional, and morphological characteristics. Second, a Hierarchical Brain Transformer Encoder processes region-specific connectivity profiles as variable-length token sequences and jointly captures local functional structure and global brain functional organization, conditioned on brain regional context, without assuming a fixed number of regions. Together, these two techniques enable the learning of generalizable representations across brain atlases with different parcellation resolutions and connectivity dimensions. AREION is pretrained on a large-scale multi-atlas cohort of fMRI data via self-supervised modeling and evaluated across diverse downstream tasks. Extensive experiments show that it outperforms strong supervised baselines and fMRI foundation models while using substantially fewer parameters, demonstrating strong transferability to atlases unseen during pretraining, generalization to external datasets, and excellent interpretability at the brain regional level.
ER -