Deep learning in digital pathology image analysis: a survey
- PDF / 651,909 Bytes
- 18 Pages / 595.276 x 785.197 pts Page_size
- 112 Downloads / 259 Views
REVIEW
Deep learning in digital pathology image analysis: a survey ✉)1,2,3,4
Shujian Deng1,2,3,*, Xin Zhang1,2,3,*, Wen Yan1,2,3, Eric I-Chao Chang4, Yubo Fan1,2,3, Maode Lai5, Yan Xu ( 1
School of Biological Science and Medical Engineering, Beihang University, Beijing 100191, China; 2Key Laboratory of Biomechanics and Mechanobiology of Ministry of Education and State Key Laboratory of Software Development Environment, Beihang University, Beijing 100191, China; 3Beijing Advanced Innovation Center for Biomedical Engineering, Beihang University, Beijing 100191, China; 4Microsoft Research Asia, Beijing 100080, China; 5Department of Pathology, School of Medicine, Zhejiang University, Hangzhou 310007, China
© Higher Education Press 2020
Abstract deep learning (DL) has achieved state-of-the-art performance in many digital pathology analysis tasks. Traditional methods usually require hand-crafted domain-specific features, and DL methods can learn representations without manually designed features. In terms of feature extraction, DL approaches are less labor intensive compared with conventional machine learning methods. In this paper, we comprehensively summarize recent DL-based image analysis studies in histopathology, including different tasks (e.g., classification, semantic segmentation, detection, and instance segmentation) and various applications (e.g., stain normalization, cell/gland/region structure analysis). DL methods can provide consistent and accurate outcomes. DL is a promising tool to assist pathologists in clinical diagnosis. Keywords
pathology; deep learning; segmentation; detection; classification
Introduction In clinical medicine, pathological examination has been regarded as a gold standard for cancer diagnosis for more than 100 years [1]. Pathologists use a microscope to observe a histological section. Many advanced technologies, including hematoxylin and eosin (H&E) staining and spectral methods, have been applied in the preparation of tissue slides to improve imaging quality. However, intraand interobserver disagreement cannot be avoided through visual observation and subjective interpretation, especially for experienced pathologists [2,3]. The limited agreement has resulted in the necessity of computational methods for pathological diagnosis [4–16] because automatic approaches can attain robust performance. The first step for computer-aided analysis is digital imaging. Digital imaging is the process of acquiring, compressing, storing, and displaying scenes digitally. Whole slide imaging is a more advanced and frequently used technology in pathology compared with traditional digital imaging technologies that process static images through cameras [17–19]. This technology involves two processes.
Received August 19, 2019; accepted March 5, 2020 Correspondence: Yan Xu, [email protected] *
Equal co-authorship, ordered alphabetically.
A specialized scanner is utilized to convert an entire glass histopathology or cytopathology slide into a digital slide. A virtual slide viewer is used to visualize the
Data Loading...
