Artificial Intelligence Applied to Histological Images of Colorectal Cancer Manabu Takamatsu 1 1Division of Pathology, the Cancer Institute, Japanese Foundation for Cancer Research, Tokyo Keyword: 大腸癌 , T1癌 , 人工知能 , 機械学習 , リンパ節転移予測 pp.495-499
Published Date 2021/4/25
DOI https://doi.org/10.11477/mf.1403202307
  • Abstract
  • Look Inside
  • Reference

 The application of AI(artificial intelligence)in the field of pathology is growing beyond mere classification of histological types to uses that contribute to treatment strategies. Neural networks and other machine learning techniques can be used to classify colorectal histological images, estimate tumor morphology that can affect prognosis, and even create models that predict lymph node metastasis of T1 colorectal cancer. It is difficult to predict prognosis with high reproducibility using the conventional histological evaluation method. However, it is possible to improve reproducibility and accuracy of prognostication by developing AI based on effective teacher data. Here I introduce the methods and discussion points for utilizing AI in the field of colorectal cancer.

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


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