Classification of histopathological whole slide images based on multiple weighted semi-supervised domain adaptation. (March 2022)

• Record Type: Journal Article
• Authors: Wang, Pin; Li, Pufei; Li, Yongming; Xu, Jin; Jiang, Mingfeng
• Is Part Of: Biomedical signal processing and control. Volume 73(2022)
• Publication Date: 2022-03
• Subjects: Histopathological images -- Deep transfer learning -- Semi-supervised domain adaptation -- Multiple weighted loss strategy -- Manifold regularization; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.103400
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: A novel deep transferred semi-supervised domain adaptation framework HisNet-SSDA is proposed for histopathological image classification with only limited labeled WSIs. A novel CNN model HisNet is designed for extracting high level features from the patches. A multiple weighted domain adaptation loss function strategy is proposed to boost the performance of HisNet-SSDA including a cross-entropy loss, a maximum mean discrepancy, an unlabeled conditional entropy loss and a novel manifold regularization term. Abstract: Deep learning has become more important in histopathological images classification for computer-aided cancer diagnosis. However, accurate histopathological image classification based on deep network relies on lots of labeled images, while the expert annotation of whole slide images (WSIs) is time-consuming and laborious. Therefore, how to obtain good classification results with limited labeled samples is still a major challenging task. To overcome the above difficulty, a deep transferred semi-supervised domain adaptation model (HisNet-SSDA) is proposed for classification of histopathological WSIs. Semi-supervised domain adaptation transfers knowledge from a label-rich source domain to a partially labeled target domain. First, a transferred pre-trained network HisNet is designed for high-level feature extraction of the randomly sampled patches from the source and target domains. Then the features of the two domains are aligned through semi-supervised domain adaptation utilizing a multiple weighted loss functions criterion which contains a novel manifold regularization term. The predicted probabilities of sampled patches are aggregated for the image-level classification. Classification results evaluated on two colon cancer datasets demonstrate the remarkable performance of the proposed method (accuracy: 94.32%±0.49%, sensitivity: 94.59%±0.46%, specificity: 94.06%±0.27% and accuracy: 91.92%±0.32%, sensitivity: 92.01%±0.47%, specificity: 91.83%±0.23%), which indicate that the proposed method can be an effective tool for WSIs classification in clinical practice.