Continual learning across population cohorts with distribution shift: insights from multi-cohort metabolic syndrome identification.

Journal: Journal of the American Medical Informatics Association : JAMIA
Published Date: Aug 1, 2025

Abstract

OBJECTIVE: This study aims to tackle the critical challenge of adapting deep learning (DL) models for deployment in real-world healthcare settings, specifically focusing on catastrophic forgetting due to distribution shifts between hospital and non-hospital environments. Metabolic syndrome (MetS) is susceptible to misdiagnosis by DL models due to distribution shifts. This work demonstrates the potential of continual learning (CL) to enhance model performance in MetS identification across diverse settings.

Authors

  • Chang Liu
    Key Lab of Cell Differentiation and Apoptosis of Ministry of Education, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
  • Zhangdaihong Liu
    Department of Engineering Science, Institute of Biomedical Engineering, University of Oxford, Oxford OX1 2JD, UK.
  • Jingjing Liu
    School of Electro-Mechanical Engineering, Xidian University, Xi'an 710071, China.
  • Chenglai Cai
    Dime Software, Wuxi, Jiangsu 214100, China.
  • David A Clifton
  • Hui Wang
    Department of Vascular Surgery, Xuanwu Hospital, Capital Medical University, Beijing, China.
  • Yang Yang
    Department of Gastrointestinal Surgery, The Third Hospital of Hebei Medical University, Shijiazhuang, China.

Keywords

External Resources

Popular Topics
Recent Journals