Back to all papers

Enhanced multimodal MRI classification of schizophrenia through cross-attention graph neural networks.

July 18, 2026pubmed logopapers

Authors

Gao J,Wu J,Qian M,Wang Z,Li Y,Luo N,Dai X,Shi W,Li P,Chen J,Chen Y,Wang H,Liu W,Li Z,Yang Y,Guo H,Wan P,Lv L,Lu L,Yan J,Song Y,Wang H,Zhang H,Du Y,Cheng Y,Xu J,Xu X,Zhang D,Jiang T

Affiliations (17)

  • School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, 611731, China.
  • School of Electrical Engineering and Electronic Information, Xihua University, Chengdu, 610039, China.
  • Brainnetome Center, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China.
  • School of Automation, Chongqing University, Chongqing, 400044, China.
  • Brainnetome Center and National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, China.
  • Institute of Mental Health, Peking University Sixth Hospital, Beijing, China; Key Laboratory of Mental Health, Ministry of Health, and National Clinical Research Center for Mental Disorders, Peking University Sixth Hospital, Beijing, China.
  • Department of Radiology, Renmin Hospital of Wuhan University, Wuhan, China.
  • Department of Psychiatry, Xijing Hospital, The Fourth Military Medical University, Xi'an, China.
  • Zhumadian Psychiatric Hospital, Zhumadian, China.
  • Department of Psychiatry, Henan Mental Hospital, The Second Affiliated Hospital of Xinxiang Medical University, Xinxiang, China.
  • Department of Psychiatry, Henan Mental Hospital, The Second Affiliated Hospital of Xinxiang Medical University, Xinxiang, China; Henan Key Lab of Biological Psychiatry of Xinxiang Medical University, International Joint Research Laboratory for Psychiatry and Neuroscience of Henan, Xinxiang, China.
  • Department of Psychiatry, Renmin Hospital of Wuhan University, Wuhan, China.
  • Department of Psychiatry, Henan Mental Hospital, The Second Affiliated Hospital of Xinxiang Medical University, Xinxiang, China; Henan Key Lab of Biological Psychiatry of Xinxiang Medical University, International Joint Research Laboratory for Psychiatry and Neuroscience of Henan, Xinxiang, China; Department of Psychology, Xinxiang Medical University, Xinxiang, China.
  • School of Computer and Information Technology, Shanxi University, Taiyuan, China.
  • Department of Psychiatry, First Affiliated Hospital of Kunming Medical University, Kunming, China.
  • Institute of Mental Health, Peking University Sixth Hospital, Beijing, China; Key Laboratory of Mental Health, Ministry of Health, and National Clinical Research Center for Mental Disorders, Peking University Sixth Hospital, Beijing, China; Center for Life Sciences/PKU-IDG/McGovern Institute for Brain Research, Peking University, Beijing, China.
  • Brainnetome Center, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China; Center for Excellence in Brain Science and Intelligence Technology, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China; Research Center for Augmented Intelligence, Zhejiang Lab, Hangzhou, 311100, China; Xiaoxiang Institute for Brain Health and Yongzhou Central Hospital, Yongzhou, 425000, Hunan Province, China. Electronic address: [email protected].

Abstract

We propose TB-GCAN, a tri-branch cross-attention graph neural network for schizophrenia classification using multimodal MRI, including sMRI, fMRI, and DTI. Built on a multi-site dataset of 1191 samples from seven scanning sites, the model exploits atlas-defined one-to-one anatomical correspondence across modalities to enable node-level cross-attention during intermediate representation learning. In 7-site leave-one-site-out evaluation, TB-GCAN achieved 84.63% accuracy and outperformed GAT, GCN, CNN, SVM, and MMGNN in the tri-modal setting. Attention-based region ranking highlighted biologically plausible schizophrenia-related regions, and downstream analyses linked the learned imaging representations to PANSS dimensions and transcriptional programs. Unlike generic multimodal GNNs that learn cross-modal relations from data, TB-GCAN directly leverages atlas-aligned regional correspondence to perform anatomically constrained node-level interaction. These findings indicate that anatomically grounded node-level multimodal fusion can improve classification performance while preserving neurobiological interpretability, thereby providing a principled framework for multimodal schizophrenia classification and biomarker discovery.

Topics

Journal Article

Ready to Sharpen Your Edge?

Subscribe to join 11k+ peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.