Affordable Access

Access to the full text

Detection of gastritis by a deep convolutional neural network from double-contrast upper gastrointestinal barium X-ray radiography

Authors
  • Togo, Ren1
  • Yamamichi, Nobutake2
  • Mabe, Katsuhiro3
  • Takahashi, Yu2
  • Takeuchi, Chihiro2
  • Kato, Mototsugu3
  • Sakamoto, Naoya4
  • Ishihara, Kenta1
  • Ogawa, Takahiro1
  • Haseyama, Miki1
  • 1 Hokkaido University, Graduate School of Information Science and Technology, N-14, W-9, Kita-ku, Sapporo, Hokkaido, 060-0814, Japan , Sapporo (Japan)
  • 2 The University of Tokyo, Department of Gastroenterology, Graduate School of Medicine, 7-3-1, Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan , Tokyo (Japan)
  • 3 National Hospital Organization Hakodate Hospital, Department of Gastroenterology, 18-16, Kawahara-cho, Hakodate City, Hokkaido, 041-8512, Japan , Hakodate City (Japan)
  • 4 Hokkaido University Graduate School of Medicine, Department of Gastroenterology, Sapporo, 060-8648, Japan , Sapporo (Japan)
Type
Published Article
Journal
Journal of Gastroenterology
Publisher
Springer Japan
Publication Date
Oct 03, 2018
Volume
54
Issue
4
Pages
321–329
Identifiers
DOI: 10.1007/s00535-018-1514-7
Source
Springer Nature
Keywords
License
Yellow

Abstract

BackgroundDeep learning has become a new trend of image recognition tasks in the field of medicine. We developed an automated gastritis detection system using double-contrast upper gastrointestinal barium X-ray radiography.MethodsA total of 6520 gastric X-ray images obtained from 815 subjects were analyzed. We designed a deep convolutional neural network (DCNN)-based gastritis detection scheme and evaluated the effectiveness of our method. The detection performance of our method was compared with that of ABC (D) stratification.ResultsSensitivity, specificity, and harmonic mean of sensitivity and specificity of our method were 0.962, 0.983, and 0.972, respectively, and those of ABC (D) stratification were 0.925, 0.998, and 0.960, respectively. Although there were 18 false negative cases in ABC (D) stratification, 14 of those 18 cases were correctly classified into the positive group by our method.ConclusionsDeep learning techniques may be effective for evaluation of gastritis/non-gastritis. Collaborative use of DCNN-based gastritis detection systems and ABC (D) stratification will provide more reliable gastric cancer risk information.

Report this publication

Statistics

Seen <100 times