Detection of multiple lesions of gastrointestinal tract for endoscopy using artificial intelligence model: a pilot study
- PDF / 971,509 Bytes
- 7 Pages / 595.276 x 790.866 pts Page_size
- 74 Downloads / 189 Views
and Other Interventional Techniques
Detection of multiple lesions of gastrointestinal tract for endoscopy using artificial intelligence model: a pilot study Linjie Guo1,2 · Hui Gong2 · Qiushi Wang3 · Qiongying Zhang2 · Huan Tong1,2 · Jing Li1,2 · Xiang Lei3 · Xue Xiao2 · Chuanhui Li2 · Jinsun Jiang4 · Bing Hu2 · Jie Song3 · Chengwei Tang1,2 · Zhiyin Huang2 Received: 10 July 2020 / Accepted: 3 November 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Background This study was aimed to develop a computer-aided diagnosis (CAD) system with deep-learning technique and to validate its efficiency on detecting the four categories of lesions such as polyps, advanced cancer, erosion/ulcer and varices at endoscopy. Methods A deep convolutional neural network (CNN) that consists of more than 50 layers were trained with a big dataset containing 327,121 white light images (WLI) of endoscopy from 117,005 cases collected from 2012 to 2017. Two CAD models were developed using images with or without annotation of the training dataset. The efficiency of the CAD system detecting the four categories of lesions was validated by another dataset containing consecutive cases from 2018 to 2019. Results A total of 1734 cases with 33,959 images were included in the validation datasets which containing lesions of polyps 1265, advanced cancer 500, erosion/ulcer 486, and varices 248. The CAD system developed in this study may detect polyps, advanced cancer, erosion/ulcer and varices as abnormality with the sensitivity of 88.3% and specificity of 90.3%, respectively, in 0.05 s. The training datasets with annotation may enhance either sensitivity or specificity about 20%, p = 0.000. The sensitivities and specificities for polyps, advanced cancer, erosion/ulcer and varices reached about 90%, respectively. The detect efficiency for the four categories of lesions reached to 89.7%. Conclusion The CAD model for detection of multiple lesions in gastrointestinal lumen would be potentially developed into a double check along with real-time assessment and interpretation of the findings encountered by the endoscopists and may be a benefit to reduce the events of missing lesions. Keywords Multiple-lesion detection · Gastrointestinal endoscopy · Deep convolutional neural network · Computer-aided diagnosis
Linjie Guo and Hui Gong have contributed equally. * Chengwei Tang [email protected] * Zhiyin Huang [email protected] 1
West China Hospital, Sichuan University, Laboratory of Gastroenterology and Hepatology, State Key Laboratory of Biotherapy, Chengdu, China
2
West China Hospital, Sichuan University, Department of Gastroenterology, Chengdu, China
3
SeedsMed Technology Inc, Sichuan, China
4
West China Hospital, Sichuan University, Management Department of Clinical Study, Chengdu, China
Gastrointestinal (GI) endoscopy is a procedure that allows for visualization of the GI lumen and determination of many pathological lesions and has been essential for the evaluation of signs and symptoms of a wide vari
Data Loading...
