CLMS: Bridging domain gaps in medical imaging segmentation with source-free continual learning for robust knowledge transfer and adaptation.

Journal: Medical image analysis
Published Date:

Abstract

Deep learning shows promise for medical image segmentation but suffers performance declines when applied to diverse healthcare sites due to data discrepancies among the different sites. Translating deep learning models to new clinical environments is challenging, especially when the original source data used for training is unavailable due to privacy restrictions. Source-free domain adaptation (SFDA) aims to adapt models to new unlabeled target domains without requiring access to the original source data. However, existing SFDA methods face challenges such as error propagation, misalignment of visual and structural features, and inability to preserve source knowledge. This paper introduces Continual Learning Multi-Scale domain adaptation (CLMS), an end-to-end SFDA framework integrating multi-scale reconstruction, continual learning, and style alignment to bridge domain gaps across medical sites using only unlabeled target data or publicly available data. Compared to the current state-of-the-art methods, CLMS consistently and significantly achieved top performance for different tasks, including prostate MRI segmentation (improved Dice of 10.87 %), colonoscopy polyp segmentation (improved Dice of 17.73 %), and plus disease classification from retinal images (improved AUC of 11.19 %). Crucially, CLMS preserved source knowledge for all the tasks, avoiding catastrophic forgetting. CLMS demonstrates a promising solution for translating deep learning models to new clinical imaging domains towards safe, reliable deployment across diverse healthcare settings.

Authors

  • Weilu Li
    Institutes of Physical Science and Information Technology, Anhui University, Hefei, 230601, China.
  • Yun Zhang
    Biology Institute, Qilu University of Technology (Shandong Academy of Sciences), Jinan, China.
  • Hao Zhou
    State Key Laboratory of Environment Health (Incubation), Key Laboratory of Environment and Health, Ministry of Education, Key Laboratory of Environment and Health (Wuhan), Ministry of Environmental Protection, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, #13 Hangkong Road, Wuhan, Hubei 430030, China.
  • Wenhan Yang
  • Zhi Xie
    State Key Lab of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou 500040, China. Electronic address: xiezh8@sysu.edu.cn.
  • Yao He
    School of Environmental Science and Engineering, Institute of Environmental Health and Pollution Control, Guangdong University of Technology, Guangzhou 510006, China.