Edge-aware Hard Clustering Graph Pooling for Brain Imaging

Graph Convolutional Networks (GCNs) can capture non-Euclidean spatial dependence between different brain regions. The graph pooling operator, a crucial element of GCNs, enhances the representation learning capability and facilitates the acquisition of abnormal brain maps. However, most existing research designs graph pooling operators solely from the perspective of nodes while disregarding the original edge features, in a way that not only confines graph pooling application scenarios, but also diminishes its ability to capture critical substructures. To design a graph clustering pooling operator that is tailored to dominant edge features, we proposed the edge-aware hard clustering graph pool (EHCPool) and redefined the graph clustering process. Specifically, the 'Edge-to-node' criterion was proposed to evaluate the significance of both edge and node features. Guided by edge scores, we designed a revolutionary Iteration n-top strategy, aimed at adaptively learning sparse hard clustering assignments for graphs. Subsequently, a novel N-E Aggregation strategy is introduced to aggregate node and edge information in each independent subgraph. Extensive experiments on the multi-site public datasets demonstrate the superiority and robustness of the proposed model. More notably, EHCPool has the potential to probe different types of dysfunctional brain networks from a data-driven perspective. Core code is at: https://github.com/swfen/EHCPool.