RestainNet: A self-supervised digital re-stainer for stain normalization. (October 2022)

• Record Type: Journal Article
• Authors: Zhao, Bingchao; Han, Chu; Pan, Xipeng; Lin, Jiatai; Yi, Zongjian; Liang, Changhong; Chen, Xin; Li, Bingbing; Qiu, Weihao; Li, Danyi; Liang, Li; Wang, Ying; Liu, Zaiyi
• Is Part Of: Computers & electrical engineering. Volume 103(2022)
• Publication Date: 2022-10
• Subjects: Computational pathology -- Stain normalization -- Self-supervised learning -- Generative adversarial networks -- Whole slide images; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108304
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Color inconsistency is an inevitable challenge in computational pathology, which harms the pathological image analysis methods, especially the learning-based models. A series of approaches have been proposed for stain normalization. However, most of them are lack of flexibility in practice. In this paper, we formulated stain normalization as a digital re-staining process and proposed a self-supervised learning model, which is called RestainNet. Our network is regarded as a digital re-stainer which learns how to re-stain an unstained (grayscale) image. Two digital stains, Hematoxylin (H) and Eosin (E), were extracted from the original image by Beer–Lambert's Law. We proposed a staining loss to maintain the correctness of stain intensity during the re-staining process. Our RestainNet outperforms existing approaches and achieves outstanding performance with regard to color correctness and structure preservation. We further conducted experiments on the segmentation and classification tasks and the proposed RestainNet achieved outstanding performance compared with SOTA methods. Graphical abstract: Highlights: Our network is trained in a self-supervised manner. Our network follows the inherent optical imaging properties of pathological images. Our network can be trained on a single Whole Slide Image. Our network achieves structure-preserving stain normalization.