Research Area:  Machine Learning

Abstract:

Manual inspection of histopathological images is important in clinical cancer diagnosis. Pathologists implement pathological diagnosis and prognostic evaluation through the microscopic examination of histopathological slices. This entire process is time-consuming, laborious, and challenging for pathologists. The modern use of whole-slide imaging, which scans histopathology slides to digital slices, and analysis using computer-aided diagnosis is an essential problem.To solve the problem of difficult labeling of histopathological data, and improve the flexibility of histopathological analysis in clinical applications, we herein propose a semi-supervised learning algorithm coupled with consistency regularization strategy, called“Semi- supervised Histopathology Analysis Network”(Semi-His-Net), for automated normal-versus-tumor and subtype classifications. Specifically, when inputted disturbing versions of the same image, the model should predict similar outputs. Based on this, the model itself can assign artificial labels to unlabeled data for subsequent model training, thereby effectively reducing the labeled data required for training.

Keywords:  
Semi-supervised learning
Tumor
Histopathological
Images analysis
Pathologists

Author(s) Name:  Yanyun Jiang, Xiaodan Sui, Yanhui Ding, Wei Xiao, Yuanjie Zheng,Yongxin Zhang

Publisher name:  Frontiers

DOI:  10.3389/fonc.2022.1044026

Volume Information:  Volume 12

Paper Link:   https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.1044026/full