Deep learning analyzes Helicobacter pylori infection by upper gastrointestinal endoscopy images

Abstract

Background and study aims Helicobacter pylori (HP)-associated chronic gastritis can cause mucosal atrophy and intestinal metaplasia, both of which increase the risk of gastric cancer. The accurate diagnosis of HP infection during routine medical checks is important. We aimed to develop a convolutional neural network (CNN), which is a machine-learning algorithm similar to deep learning, capable of recognizing specific features of gastric endoscopy images. The goal behind developing such a system was to detect HP infection early, thus preventing gastric cancer.

Patients and methods For the development of the CNN, we used 179 upper gastrointestinal endoscopy images obtained from 139 patients (65 were HP-positive: ≥ 10 U/mL and 74 were HP-negative: < 3 U/mL on HP IgG antibody assessment). Of the 179 images, 149 were used as training images, and the remaining 30 (15 from HP-negative patients and 15 from HP-positive patients) were set aside to be used as test images. The 149 training images were subjected to data augmentation, which yielded 596 images. We used the CNN to create a learning tool that would recognize HP infection and assessed the decision accuracy of the CNN with the 30 test images by calculating the sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC).

Results The sensitivity and specificity of the CNN for the detection of HP infection were 86.7 % and 86.7 %, respectively, and the AUC was 0.956.

Conclusions CNN-aided diagnosis of HP infection seems feasible and is expected to facilitate and improve diagnosis during health check-ups.

• References

• 1 Tahara T, Arisawa T, Shibata T. et al. Risk prediction of gastric cancer by analysis of aberrant DNA methylation in non-neoplastic gastric epithelium. Digestion 2007; 75: 54-61
• 2 Uemura N, Okamoto S, Yamamoto S. et al. Helicobacter pylori infection and the development of gastric cancer. N Engl J Med 2001; 345: 784-789
• 3 Goodwin CS. Helicobacter pylori gastritis, peptic ulcer, and gastric cancer: clinical and molecular aspects. Clin Infect Dis 1997; 25: 1017-1019
• 4 Nomura S, Terao S, Adachi K. et al. Endoscopic diagnosis of gastric mucosal activity and inflammation. Dig Endosc 2013; 25: 136-146
• 5 Tahara T, Shibata T, Horiguchi N. et al. A possible link between gastric mucosal atrophy and gastric cancer after Helicobacter pylori eradication. PLoS One 2016; 11: e0163700
• 6 Laine L, Cohen H, Sloane R. et al. Interobserver agreement and predictive value of endoscopic findings for H. pylori and gastritis in normal volunteers. Gastrointest Endosc 1995; 42: 420-423
• 7 Watanabe K, Nagata N, Shimbo T. et al. Accuracy of endoscopic diagnosis of Helicobacter pylori infection according to level of endoscopic experience and the effect of training. BMC Gastroenterol 2013; 13: 128-134
• 8 Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Adv Neural Inf Process Syst (NIPS Proceedings) 2012; 1106-1114
• 9 Jia Y, Shelhamer E, Donahue J. et al. Caffe: convolutional architecture for fast feature embedding. Proc ACM Int Conf Multimed 2014; 675-678
• 10 Mori Y, Kudo S, Wakamura K. et al. Novel computer-aided diagnostic system for colorectal lesions by using endocytoscopy (with videos). Gastrointest Endosc 2015; 81: 621-629
• 11 Kimura K, Takemoto T. An endoscopic recognition of the atrophic border and its significance in chronic gastritis. Endoscopy 1969; 1: 87-97
• 12 Szegedy C, Liu W, Jia Y. et al. Going deeper with convolutions. 2015 IEEE Conf Comput Vis Pattern Recognit 2015; 1-9
• 13 Deng J, Dong W, Socher R. et al. ImageNet: a large-scale hierarchical image database. 2009 IEEE Conf Comput Vis Pattern Recognit 2009; 2-9
• 14 Russakovsky O, Deng J, Su H. et al. ImageNet Large Scale Visual Recognition Challenge. Int J Comput Vis 2015; 115: 211-252
• 15 Girshick R, Donahue J, Darrell T. et al. Rich feature hierarchies for accurate object detection and semantic segmentation. Proc IEEE Comput Conf Comput Vis Pattern Recognit 2014; 580-587
• 16 Hinton GE, Salakhutdinov RR. Reducing the dimensionality of data with neural networks. Science 2006; 313: 504-507
• 17 Bottou L. Large-scale machine learning with stochastic gradient descent. Proc COMPSTAT’2010 2010; 177-186
• 18 Bornschein J, Malfertheiner P. Gastric carcinogenesis. Langenbecks Arch Surg 2011; 396: 729-742
• 19 Asaka M, Sugiyama T, Nobuta A. et al. Atrophic gastritis and intestinal metaplasia in Japan: results of a large multicenter study. Helicobacter 2001; 6: 294-299
• 20 Sugano K, Tack J, Kuipers EJ. et al. faculty members of Kyoto Global Consensus Conference. Kyoto global consensus report on Helicobacter pylori gastritis. Gut 2015; 64: 1353-1367
• 21 Ji R, Li YQ. Diagnosing Helicobacter pylori infection in vivo by novel endoscopic techniques. World J Gastroenterol 2014; 20: 9314-9320
• 22 Okubo M, Tahara T, Shibata T. et al. Usefulness of magnifying narrow-band imaging endoscopy in the Helicobacter pylori-related chronic gastritis. Digestion 2011; 83: 161-166
• 23 Nakagawa S, Kato M, Shimizu Y. et al. Relationship between histopathologic gastritis and mucosal microvascularity: Observations with magnifying endoscopy. Gastrointest Endosc 2003; 58: 71-75
• 24 Tahara T, Shibata T, Nakamura M. et al. Gastric mucosal pattern by using magnifying narrow-band imaging endoscopy clearly distinguishes histological and serological severity of chronic gastritis. Gastrointest Endosc 2009; 70: 246-253
• 25 Lee S-Y. Current progress toward eradicating Helicobacter pylori in East Asian countries: differences in the 2013 revised guidelines between China, Japan, and South Korea. World J Gastroenterol 2014; 20: 1493-1502
• 26 Di Rienzo TA, D’Angelo G, Ojetti V. et al. 13C-Urea breath test for the diagnosis of Helicobacter pylori infection. Eur Rev Med Pharmacol Sci 2013; 17: 51-58
• 27 Kato M, Asaka M, Saito M. et al. Clinical usefulness of urine-based enzyme-linked immunosorbent assay for detection of antibody to Helicobacter pylori: a collaborative study in nine medical institutions in Japan. Helicobacter 2000; 5: 109-119
• 28 Bah A, Saraga E, Armstrong D. et al. Endoscopic Features of Helicobacter pylori-related gastritis. Endoscopy 1995; 27: 59-56
• 29 Naylor GM, Gotoda T, Dixon M. et al. Why does Japan have a high incidence of gastric cancer? Comparison of gastritis between UK and Japanese patiets. Gut 2006; 55: 1545-1552