迈向多视图脑网络基础模型：跨任意脑图谱的跨视图一致性学习
Toward a Multi-View Brain Network Foundation Model: Cross-View Consistency Learning across Arbitrary Atlases
脑网络分析可以帮助理解大脑组织方式，也常用于识别神经系统疾病，但现有模型往往依赖某一种脑图谱划分，难以迁移。作者提出 MV-BrainFM，一个多视图脑网络基础模型，可在任意脑图谱构建的脑网络上学习通用表征。它把解剖距离信息引入 Transformer（一种注意力机制模型）来引导脑区之间的交互，并用无监督的跨视图一致性学习，把同一受试者不同脑图谱下的表征对齐到同一潜在空间。模型还支持多数据集、多图谱同时预训练，计算效率优于传统顺序训练。在来自 17 个功能磁共振数据集、超过两万名受试者上的实验中，该模型在单图谱和多图谱设置下均优于 14 个现有脑网络基础模型和任务专用基线，学到的表征在解剖和临床上具有可解释性。
为什么推荐给您：方法学创新且规模大，跨图谱泛化有转化潜力，但属计算模型研究尚未进入临床。
不需要生物学背景，多打比方
正在获取全文并生成讲解（拿不到全文就依据摘要），大约需要 30–60 秒…
已等待 0 秒
这篇还没有动画
动画会把研究的流程、作用机制和关键结果一步一步演示出来，每一步都标明出自原文哪里。制作大约需要 30–60 秒。
摘要Abstract
Brain network analysis provides an interpretable framework for characterizing brain organization and has been widely used for neurological disorder identification. Recent advances in self-supervised learning have motivated the development of brain network foundation models. However, existing approaches are often limited by atlas dependency, insufficient exploitation of multiple network views, and weak incorporation of anatomical priors. In this work, we propose MV-BrainFM, a multi-view brain network foundation model designed to learn generalizable and scalable representations from brain networks constructed with arbitrary atlases. MV-BrainFM explicitly in-corporates anatomical distance information into Transformer based modeling to guide inter-regional interactions, and introduces an unsupervised cross-view consistency learning strategy to align representations from multiple atlases of the same subject in a shared latent space. By jointly enforcing within view robustness and cross-view alignment during pretraining, the model effectively captures complementary information across heterogeneous network views while remaining atlas-aware. In addition, MV-BrainFM adopts a unified multi-view pretraining paradigm that enables simultaneous learning from multiple datasets and atlases, significantly improving computational efficiency compared to conventional sequential training strategies. The proposed framework also demonstrates strong scalability, consistently benefiting from increasing data diversity while maintaining stable performance across unseen atlas configurations. Extensive experiments on more than 20,000 subjects from 17 fMRI datasets show that MV-BrainFM consistently outperforms 14 existing brain network foundation models and task-specific baselines under both single-atlas and multi-atlas settings. Furthermore, the learned representations exhibit strong anatomical and clinical interpretability, highlighting meaningful brain regions and cross-view consistent patterns associated with neurological conditions. These results suggest that MV-BrainFM provides an effective, efficient, and scalable foundation model for brain network analysis. The source code is available at https://github.com/AngusMonroe/MV-BrainFM.