Deep texture representation analysis for histopathological images

Ranny Rahaningrum Herdiantoputri; Daisuke Komura; Kei Fujisaka; Tohru Ikeda; Shumpei Ishikawa

doi:10.1016/j.xpro.2023.102161

Deep texture representation analysis for histopathological images

STAR Protoc. 2023 Mar 23;4(2):102161. doi: 10.1016/j.xpro.2023.102161. Online ahead of print.

Authors

Ranny Rahaningrum Herdiantoputri¹, Daisuke Komura², Kei Fujisaka³, Tohru Ikeda¹, Shumpei Ishikawa⁴

Affiliations

¹ Department of Oral Pathology, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, 13 1-5-45, Yushima, Bunkyo-ku, Tokyo 1138549, Japan.
² Department of Preventive Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 1130033, Japan. Electronic address: kdais-prm@m.u-tokyo.ac.jp.
³ Department of Preventive Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 1130033, Japan.
⁴ Department of Preventive Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 1130033, Japan; Division of Pathology, National Cancer Center Exploratory Oncology Research & Clinical Trial Center, 6-5-1, Kashiwanoha, Kashiwa, Chiba 2778577, Japan. Electronic address: ishum-prm@m.u-tokyo.ac.jp.

Abstract

Deep texture representations (DTRs) produced from a bilinear convolutional neural network allow objective quantification of tumor histopathology images effectively. They can be used for various analyses, including visualization of morphological correlation between histology images, content-based image retrieval (CBIR), and supervised learning. This protocol describes the simplified workflow to analyze DTRs from data preparation, visualization of the histological profile, and CBIR analysis, to supervised learning model development to predict the profile from histological images. For complete details on the use and execution of this protocol, please refer to Komura et al. (2022).¹.

Keywords: Cancer; Computer Sciences; Microscopy.