HGMA-Net: a hierarchical graph–mamba encoder for multi-modal brain disorder classification on population graphs
Biomedical Physics & Engineering Express · 2026
DOI: 10.1088/2057-1976/aeaaec ↗Kurz gefasst
Fachartikel in der Zeitschrift Biomedical Physics & Engineering Express, 2026. Verfasst von Wanting Lai und 5 weiteren Autor:innen. Beteiligte Einrichtung: Guangdong Provincial Engineering and Technology Research Center of Light and Health, Guangdong Pharmaceutical University, Guangzhou 510006, People's …. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Brain Diseases, Brain, Image Processing, Computer-Assisted. Die Arbeit nennt 28 Literaturangaben.
Abstract (Originalsprache)
Brain disorder (BD) classification based on population graph learning has shown promising performance by jointly modeling resting-state functional magnetic resonance imaging data and phenotypic information. Existing graph Transformer–based population graph encoders suffer from high computational cost on densely connected subject graphs and lack an efficient mechanism for global contextual modeling across distant subjects. To address these limitations, we propose HGMA-Net, a hierarchical graph–Mamba hybrid framework for multi-modal BD classification. Specifically, the proposed HGMA module redesigns the shallow encoder stage using a dual-path architecture that combines local graph aggregation with global sequence modeling. A degree-guided node serialization strategy provides a topology-aware ordering of graph nodes, enabling bidirectional Mamba state space models to efficiently capture global contextual information with linear sequence complexity. An adaptive gating mechanism is further introduced to dynamically fuse global sequential features and local graph-structural features. To balance computational efficiency and representational capacity, HGMA is applied only at the first encoder layer, while deeper layers retain TransformerConv for local topology refinement. Experiments on the ABIDE and ADHD-200 datasets demonstrate that HGMA-Net achieves superior classification performance, obtaining AUC scores of 89.14% and 91.70%, respectively. HGMA-Net also retains efficient inference when global contextual modeling is introduced at the shallow encoder stage.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Guangdong Provincial Engineering and Technology Research Center of Light and Health, Guangdong Pharmaceutical University, Guangzhou 510006, People's Republic of China.
- Shenzhen Bao'an District Songgang People's Hospital, Shenzhen 518105, People's Republic of China.
ORCID-Profile: Wanting Lai
Zitieren
APA
Lai, W., Zheng, K., He, Y., Lin, X., Liu, Y., & Qiu, X. (2026). HGMA-Net: a hierarchical graph–mamba encoder for multi-modal brain disorder classification on population graphs. Biomedical Physics & Engineering Express. https://doi.org/10.1088/2057-1976/aeaaec
Vancouver
Lai W, Zheng K, He Y, Lin X, Liu Y, Qiu X. HGMA-Net: a hierarchical graph–mamba encoder for multi-modal brain disorder classification on population graphs. Biomedical Physics & Engineering Express. 2026. doi:10.1088/2057-1976/aeaaec
BibTeX
@article{Lai2026,
author = {Lai, Wanting and Zheng, Kehong and He, Yongling and Lin, Xiaozhen and Liu, Ying and Qiu, Xuejun},
title = {HGMA-Net: a hierarchical graph–mamba encoder for multi-modal brain disorder classification on population graphs},
journal = {Biomedical Physics & Engineering Express},
year = {2026},
doi = {10.1088/2057-1976/aeaaec}
}
EUVIMED