BrainLinear：用于稀疏切空间中脑网络分析的线性模型
BrainLinear: A Linear Model for Brain Network Analysis in Sparse Tangent Subspaces
- Guangdong University of Foreign Studies（广东外语外贸大学）
- The University of Tokyo（东京大学）
- RIKEN AIP（理化学研究所先进智能项目中心）
- Royal Melbourne Institute of Technology（皇家墨尔本理工大学）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
本研究针对脑网络分析中现有方法的冗余与高成本问题，提出轻量几何感知框架BrainLinear，在ABIDE和ADNI数据集上以远低于GNN、Transformer的成本达到更优性能，且支持连接层面解释。
中文摘要 AI 辅助
功能连接组分析通过研究脑区间的相互作用，以理解和识别自闭症谱系障碍、阿尔茨海默病等疾病。现有方法通常使用图神经网络（GNN）和Transformer对完整的功能连接矩阵进行建模，但处理数万个连接会引入冗余和噪声，增加计算成本，并限制连接层面的可解释性。这引发了一个核心问题：我们真的需要复杂的交互建模吗？还是只需识别少量与疾病相关的连接模式就足够了？为回答该问题，我们提出BrainLinear，一种用于挖掘疾病判别性连接组模式的轻量型几何感知框架。BrainLinear首先将每个功能连接矩阵映射到以训练集Fréchet均值为中心的共享切空间，在保留矩阵几何特性的同时捕获被试特异性偏差；接着，通过分类贡献和疾病-对照差异对每个感兴趣区域（ROI）对的切方向进行评分，保留Top-K个方向作为紧凑表示；最后，用一个浅层多层感知机对所选表示进行分类。在ABIDE和ADNI数据集上的实验表明，BrainLinear在成本仅为强大的GNN和Transformer基线的一小部分的情况下，性能与这些基线相当或更优：它在每个指标上较最佳基线的AUC和ACC分别提升了最高3.54和1.39个百分点，同时在AUC指标下，与最接近的基线相比，运行时间和峰值GPU内存分别减少了84.0%和68.4%。所选方向与组间位移方向一致，并跨主要功能系统组织，支持连接层面的解释。
英文摘要
Functional connectome analysis examines brain-region interactions to understand and identify disorders such as autism spectrum disorder and Alzheimer's disease. Existing methods typically use GNNs and Transformers to model the full functional connectivity matrix. However, processing tens of thousands of connections introduces redundancy and noise, increases computational cost, and limits connection-level interpretability. This raises a central question: do we really need complex interaction modeling, or is identifying a small set of disease-relevant connectivity patterns sufficient? To answer this question, we propose BrainLinear, a lightweight geometry-aware framework for mining disease-discriminative connectome patterns. BrainLinear first maps each functional connectivity matrix to a shared tangent space centered at the Fréchet mean of the training set, capturing subject-specific deviations while respecting matrix geometry. It then scores each ROI-pair tangent direction by its classification contribution and disease--control difference, retaining Top-$K$ directions as a compact representation. Finally, a shallow multilayer perceptron performs classification on the selected representation. Experiments on ABIDE and ADNI show that BrainLinear matches or exceeds strong GNN and Transformer baselines at a fraction of their cost: it improves AUC and ACC over the best baseline for each metric by up to $3.54$ and $1.39$ percentage points, while reducing runtime and peak GPU memory by $84.0\%$ and $68.4\%$ relative to the closest baseline in AUC. The selected directions are directionally consistent with between-group displacements and organized across major functional systems, supporting connection-level interpretation.
