Attempt to Use a Deep Learning-based Artificial Intelligence Diagnostic Algorithm for Detecting Lymph Node Metastasis of Gastric Cancer Jun Matsushima 1,2,3 , Tamotsu Sato 4 , Takashi Ohnishi 5 , Yuichiro Yoshimura 2,6 , Shinichiro Koto 4 , Hiroyuki Mizutani 4 , Shinichi Ban 1 , Jun-ichiro Ikeda 3 , Masayuki Kano 7 , Hisahiro Matsubara 7 , Hideki Hayashi 2,7 1Department of Pathology, Saitama Medical Center, Dokkyo Medical University, Saitama, Japan 2Center for Frontier Medical Engineering, Chiba University, Chiba, Japan 3Department of Diagnostic Pathology, Graduate School of Medicine, Chiba University, Chiba, Japan 4Toshiba Digital Solutions Corporation, Kawasaki, Japan 5Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, USA 6Toyama University Hospital, Toyama, Japan 7Department of Frontier Surgery, Graduate School of Medicine, Chiba University, Chiba, Japan Keyword: 胃癌 , リンパ節転移 , 病理診断 , AI , 深層学習 pp.491-494
Published Date 2021/4/25
DOI https://doi.org/10.11477/mf.1403202306
  • Abstract
  • Look Inside
  • Reference

 Recently, AI(artificial intelligence)based on deep learning has demonstrated excellent outcomes in various automated image-recognition algorithms. Our study on the performances of deep learning-based algorithm for the detection of gastric cancer metastasis in hematoxylin and eosin-stained tissue sections of lymph nodes achieved an area under the receiver operating characteristic curve(AUC)of >0.99. Similar AI algorithms to detect breast cancer metastasis in lymph nodes or to classify adenocarcinoma, adenoma, and non-neoplastic tissues in gastrointestinal biopsy histology have been studied and showed good applicability. However, some challenges exist in the introduction of AI-based pathological diagnosis in daily practice.

Copyright © 2021, Igaku-Shoin Ltd. All rights reserved.


電子版ISSN 1882-1219 印刷版ISSN 0536-2180 医学書院