Confidence-guided mask learning for semi-supervised medical image segmentation.

Journal: Computers in biology and medicine
PMID:

Abstract

Semi-supervised learning aims to train a high-performance model with a minority of labeled data and a majority of unlabeled data. Existing methods mostly adopt the mechanism of prediction task to obtain precise segmentation maps with the constraints of consistency or pseudo-labels, whereas the mechanism usually fails to overcome confirmation bias. To address this issue, in this paper, we propose a novel Confidence-Guided Mask Learning (CGML) for semi-supervised medical image segmentation. Specifically, on the basis of the prediction task, we further introduce an auxiliary generation task with mask learning, which intends to reconstruct the masked images for extremely facilitating the model capability of learning feature representations. Moreover, a confidence-guided masking strategy is developed to enhance model discrimination in uncertain regions. Besides, we introduce a triple-consistency loss to enforce a consistent prediction of the masked unlabeled image, original unlabeled image and reconstructed unlabeled image for generating more reliable results. Extensive experiments on two datasets demonstrate that our proposed method achieves remarkable performance.

Authors

  • Wenxue Li
  • Wei Lu
    Department of Pharmacy, Taihe Hospital, Hubei University of Medicine, Shiyan, China.
  • Jinghui Chu
    School of Electronic Information Engineering, Tianjin University, Tianjin 300072, PR China. Electronic address: cjh@tju.edu.cn.
  • Qi Tian
    College of Biomedical Engineering and Instrument Science, Zhejiang University, Zheda Road, 310027 Hanghzou, China; Key Laboratory for Biomedical Engineering, Ministry of Education, China. Electronic address: Tianq@zju.edu.cn.
  • Fugui Fan
    The School of Electrical and Information Engineering, Tianjin University, Tianjin, 300072, China. Electronic address: tjufan@tju.edu.cn.