Detection of gastritis by a deep convolutional neural network from double-contrast upper gastrointestinal barium X-ray radiography
Deep 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.
A 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.
Sensitivity, 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.
Deep 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.
KeywordsDeep convolutional neural network Artificial intelligence Gastritis Double-contrast upper gastrointestinal barium X-ray radiography
Deep convolutional neural network
- H. pylori
- ROC curve
Receiver operating characteristic curve
Region of interest
Upper gastrointestinal endoscopy
Double-contrast upper gastrointestinal barium X-ray radiography
The clinical data were acquired at The University of Tokyo Hospital in Japan. This study was partly supported by Global Station for Big Data and Cybersecurity, a project of Global Institution for Collaborative Research and Education at Hokkaido University JSPS KAKENHI Grant number JP17H01744.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest
- 7.Miura K, Okada H, Kouno Y, et al. Actual status of involvement of Helicobacter pylori infection that developed gastric cancer from group A of ABC (D) stratification—study of early gastric cancer cases that underwent endoscopic submucosal dissection. Digestion Karger Publishers. 2016;94:17–23.CrossRefGoogle Scholar
- 8.Oshima A, Hirata N, Ubukata T, et al. Evaluation of a mass screening program for stomach cancer with a case–control study design. Int J Cancer. Wiley Subscription Services, Inc., A Wiley Company. 1986;38:829–33.Google Scholar
- 9.Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. In: Proceedings of the 25th international conference on neural information processing systems. Lake Tahoe, Nevada: Curran Associates Inc.; 2012. pp. 1097–105.Google Scholar
- 12.Yamamichi N, Hirano C, Takahashi Y, et al. Comparative analysis of upper gastrointestinal endoscopy, double-contrast upper gastrointestinal barium X-ray radiography, and the titer of serum anti-Helicobacter pylori IgG focusing on the diagnosis of atrophic gastritis. Gastric Cancer. Springer Japan. 2016;19:670–5.CrossRefGoogle Scholar
- 15.Kimura K, Takemoto T. An endoscopic recognition of the atrophic border and its significance in chronic gastritis. Endoscopy. © Georg Thieme Verlag, Stuttgart. 1969;1:87–97.Google Scholar
- 16.Kimura K. Chronological transition of the fundic-pyloric border determined by stepwise biopsy of the lesser and greater curvatures of the stomach. Gastroenterology. 1972;63:584–92.Google Scholar
- 17.Cortes C. Support-vector networks. Mach Learn Springer. 1995;20:273–97.Google Scholar
- 19.Venegas-Barrera CS, Manjarrez J. Visual categorization with bags of keypoints. In: Workshop on statistical learning in computer vision, ECCV. 2004.Google Scholar
- 20.Jia Y, Shelhamer E, Caffe DJ. Caffe: convolutional architecture for fast feature embedding. In: Proceedings of the 22nd ACM international conference on multimedia. Orlando, Florida, USA: ACM; 2014. pp. 675–8.Google Scholar
- 23.Lakhani P, Sundaram B. Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology. Radiological Society of North America. 2017;284:574–82.Google Scholar
- 24.Kim KH, Choi SH, Park S-H. Improving arterial spin labeling by using deep learning. Radiology. Radiological Society of North America. 2018;287:658–66.Google Scholar
- 25.Larson DB, Chen MC, Lungren MP, et al. Performance of a deep-learning neural network model in assessing skeletal maturity on pediatric hand radiographs. Radiology. Radiological Society of North America. 2018;287:313–22.Google Scholar