期刊
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
卷 27, 期 1, 页码 97-108出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JBHI.2022.3216293
关键词
Image segmentation; Semantics; Pathology; Training; Task analysis; Data models; Adaptation models; Pathological image analysis; semi-supervised learning; contrastive learning
Accurate tissue segmentation in histopathological images is crucial for advancing precision pathology. We propose a semi-supervised pixel contrastive learning framework (SSPCL) to mimic the pathologist's diagnosis process and model the semantic relation of the whole slide image. SSPCL includes an uncertainty-guided mutual dual consistency learning module (UMDC) and a cross image pixel-contrastive learning module (CIPC). Experimental results show that SSPCL significantly reduces labeling cost and achieves accurate quantitation of tissues, outperforming state-of-the-art methods by up to 5.0% in mDice.
Accurate tissue segmentation in histopathological images is essential for promoting the development of precision pathology. However, the size of the digital pathological image is great, which needs to be tiled into small patches containing limited semantic information. To imitate the pathologist's diagnosis process and model the semantic relation of the whole slide image, We propose a semi-supervised pixel contrastive learning framework (SSPCL) which mainly includes an uncertainty-guided mutual dual consistency learning module (UMDC) and a cross image pixel-contrastive learning module (CIPC). The UMDC module enables efficient learning from unlabeled data through mutual dual-consistency and consensus-based uncertainty. The CIPC module aims at capturing the cross-patch semantic relationship by optimizing a contrastive loss between pixel embeddings. We also propose several novel domain-related sampling methods by utilizing the continuous spatial structure of adjacent image patches, which can avoid the problem of false sampling and improve the training efficiency. In this way, SSPCL significantly reduces the labeling cost on histopathological images and realizes the accurate quantitation of tissues. Extensive experiments on three tissue segmentation datasets demonstrate the effectiveness of SSPCL, which outperforms state-of-the-art up to 5.0% in mDice.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据