Al Alcmaeon, a three-dimensional brain MRI foundation model pretrained without manual labels on more than 425,000 volumes and derived imaging maps, can be expanded sequentially across clinical domains, and Graph-Blueprint Pruning showed less forgetting than sequential adaptation and elastic weight consolidation across voxel-level reconstruction measures.
Abstract
Brain magnetic resonance imaging (MRI) is central to neuroscience and clinical assessment, but models are commonly developed for individual diseases, populations or imaging protocols. Foundation models promise more general representations, yet they are usually pretrained once and can lose earlier capabilities when updated with new data. Here we show that Alcmaeon, a three-dimensional brain MRI foundation model pretrained without manual labels on more than 425,000 volumes and derived imaging maps, can be expanded sequentially across clinical domains. Alcmaeon combines volumetric encoding and latent diffusion generation with Graph-Blueprint Pruning (GBP), which protects network modules important to earlier domains while leaving the remaining capacity trainable. Across expansion from healthy ageing and neurodegeneration to developmental, psychiatric and tumour imaging, GBP showed less forgetting than sequential adaptation and elastic weight consolidation across voxel-level reconstruction measures, with its largest advantage after adaptation to tumour imaging. The blueprints provided an inspectable record of how model capacity was protected and reused. Representations from different model levels supported image synthesis, disease classification, survival modelling and postoperative prediction, although no single representation was optimal for every task. These findings provide a route towards brain MRI foundation models that can grow with emerging data while retaining earlier capabilities.
It is suggested that these frozen neuroimaging foundation models are insufficient for fine-grained diagnostic classification in small, non-western clinical cohorts, motivating parameter-efficient adaptation and broader multi-site external validation for equitable deployment in global health settings.
O. I. Akinmuleya, O. Akano, Samuel Danquah Ankapong et al.· 0 citations
Multiscale Volumetric Reduction (MVR), a reduce-then-encode approach that compresses each anatomical view from (D) slices into (M<<D) complementary 2D components before foundation-model encoding, achieves strong overall performance across ADNI, OASIS, and ABIDE relative to the evaluated 2D-to-3D adaptation methods and...
De-Xuan Ding, Yuan-Kai Qi, Bo-Gong Wang et al.· arXiv.org· 0 citations
BrainFedFM, a structural brain MRI foundation model federatively pretrained on 164,707 three-dimensional scans drawn from diverse real-world data distributions and organized across 42 federated sites, demonstrates the generalizability of BrainFedFM and highlights federated pretraining as a practical strategy for develo...
Zhennaan Yu, Yang Liu, Xia-Hai Zhuang et al.· 0 citations
AREION, atlas-agnostic representation learning of brain connectomes, 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 interpretabili...
Diffusion MRI (dMRI) probes water diffusion within the brain, providing unique information about tissue microstructure and white matter connectivity. High-quality dMRI enables advanced and accurate analyses but demands dense acquisitions, resulting in long scan times and hardware constraints that are often infeasible....
Xi Zhu, Yi-Jie Li, Xiao-Fan Wang et al.· 0 citations
This work presents BrainDiff, the first longitudinal vision-language system for brain MRI, which outperforms both frontier general-purpose and single-study neuroimaging models on the same patient pairs and provides a factorial over prior-report availability and image identity.
Krish Patel, Peirong Liu· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.