Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN—An Observational Study from a Tertiary Care Center

Abstract

Background and Aims Precise optical diagnosis of colorectal polyps could improve the cost-effectiveness of colonoscopy and reduce polypectomy-related complications. We conducted this study to estimate the diagnostic performance of visual inspection alone (WLI + NBI) and of EndoBRAIN (endocytoscopy-computer-aided diagnosis [EC-CAD]) in identifying a lesion as neoplastic or nonneoplastic using EC in real-world scenario.

Methods In this observational, prospective, pilot study, a total of 55 polyps were studied in the patients aged more than or equal to 18 years. EndoBRAIN is an artificial intelligence (AI)-based system that analyzes cell nuclei, crypt structure, and vessel pattern in differentiating neoplastic and nonneoplastic lesion in real-time. Endoscopist assessed polyps using white light imaging (WLI), narrow band imaging (NBI) initially followed by assessment using EC with NBI and EC with methylene blue staining. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of endoscopist and EndoBRAIN in identifying the neoplastic from nonneoplastic polyp was compared using histopathology as gold-standard.

Results A total of 55 polyps were studied, in which most of them were diminutive (36/55) and located in rectum (21/55). The image acquisition rate was 78% (43/55) and histopathology of the majority was identified to be hyperplastic (20/43) and low-grade adenoma (16/43). EndoBRAIN identified colonic polyps with 100% sensitivity, 81.82% specificity (95% confidence interval [CI], 59.7–94.8%), 90.7% accuracy (95% CI, 77.86–97.41%), 84% positive predictive value (95% CI, 68.4–92.72%), and 100% negative predictive value. The sensitivity and negative predictive value were significantly greater than visual inspection of endoscopist. The diagnostic accuracy seems to be superior; however, it did not reach statistical significance. Specificity and positive predictive value were similar in both groups.

Conclusion Optical diagnosis using EC and EC-CAD has a potential role in predicting the histopathological diagnosis. The diagnostic performance of CAD seems to be better than endoscopist using EC for predicting neoplastic lesions.

Keywords

EndoBRAIN - endocytoscopy - polyp

Publication History

Article published online:
23 December 2022

• References

• 1 Three-Year Report of Population Based Cancer Registries Incidence and Distribution of Cancer (Report of 25 PBCRs in India). 2009 [cited 2022 Jan 10]; Accessed October 15, 2022, at: www.ncrpindia.org
• 2 Gao J, Jiang Q, Zhou B, Chen D. Convolutional neural networks for computer-aided detection or diagnosis in medical image analysis: an overview. Math Biosci Eng 2019; 16 (06) 6536-6561
• 3 Ebigbo A, Palm C, Probst A. et al. A technical review of artificial intelligence as applied to gastrointestinal endoscopy: clarifying the terminology. Endosc Int Open 2019; 7 (12) E1616-E1623
• 4 Rex DK, Kahi C, O'Brien M. et al. The American Society for Gastrointestinal Endoscopy PIVI (Preservation and Incorporation of Valuable Endoscopic Innovations) on real-time endoscopic assessment of the histology of diminutive colorectal polyps. Gastrointest Endosc 2011; 73 (03) 419-422
• 5 Mori Y, Kudo SE, Wakamura K. et al. Novel computer-aided diagnostic system for colorectal lesions by using endocytoscopy (with videos). Gastrointest Endosc 2015; 81 (03) 621-629
• 6 Misawa M, Kudo SE, Mori Y. et al. Characterization of colorectal lesions using a computer-aided diagnostic system for narrow-band imaging endocytoscopy. Gastroenterology 2016; 150 (07) 1531-1532 .e3
• 7 Kudo SE, Misawa M, Wada Y. et al. Endocytoscopic microvasculature evaluation is a reliable new diagnostic method for colorectal lesions (with video). Gastrointest Endosc 2015; 82 (05) 912-923
• 8 Kudo SE, Wakamura K, Ikehara N, Mori Y, Inoue H, Hamatani S. Diagnosis of colorectal lesions with a novel endocytoscopic classification - a pilot study. Endoscopy 2011; 43 (10) 869-875
• 9 Mori Y, Kudo SE, East JE. et al. Cost savings in colonoscopy with artificial intelligence-aided polyp diagnosis: an add-on analysis of a clinical trial (with video). Gastrointest Endosc 2020; 92 (04) 905-911 .e1
• 10 Corley DA, Jensen CD, Marks AR. et al. Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med 2014; 370 (14) 1298-1306
• 11 Nazarian S, Glover B, Ashrafian H, Darzi A, Teare J. Diagnostic accuracy of artificial intelligence and computer-aided diagnosis for the detection and characterization of colorectal polyps: systematic review and meta-analysis. J Med Internet Res 2021; 23 (07) e27370
• 12 Choi SJ, Kim ES, Choi K. Prediction of the histology of colorectal neoplasm in white light colonoscopic images using deep learning algorithms. Sci Rep 2021; 11 (01) 5311
• 13 Gross S, Trautwein C, Behrens A. et al. Computer-based classification of small colorectal polyps by using narrow-band imaging with optical magnification. Gastrointest Endosc 2011; 74 (06) 1354-1359
• 14 Kudo SE, Misawa M, Mori Y. et al. Artificial intelligence-assisted system improves endoscopic identification of colorectal neoplasms. Clin Gastroenterol Hepatol 2020; 18 (08) 1874-1881 .e2
• 15 Mori Y, Kudo SE, Misawa M. et al. Artificial intelligence-assisted colonic endocytoscopy for cancer recognition: a multicenter study. Endosc Int Open 2021; 9 (07) E1004-E1011
• 16 Mori Y, Neumann H, Misawa M, Kudo SE, Bretthauer M. Artificial intelligence in colonoscopy - now on the market. What's next?. J Gastroenterol Hepatol 2021; 36 (01) 7-11
• 17 Mori Y, Kudo SE, Misawa M. et al. Real-time use of artificial intelligence in identification of diminutive polyps during colonoscopy: a prospective study. Ann Intern Med 2018; 169 (06) 357-366
• 18 Jin EH, Lee D, Bae JH. et al. Improved accuracy in optical diagnosis of colorectal polyps using convolutional neural networks with visual explanations. Gastroenterology 2020; 158 (08) 2169-2179 .e8