Hypergraph-based semantic and topological self-supervised learning for brain disease diagnosis

Open Webpage

Xiangmin Han, Mengqi Lei, Junchang Li

Published: 2026, Last Modified: 12 Nov 2025Pattern Recognit. 2026EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•HGST leverages hypergraph to learn high-order topological and semantic information.•It performs SSL by integrating node restoration and hyperedge structure similarity.•The identified high-order structure modes validate the interpretability of HGST.