Application of an artificial intelligence system for endoscopic diagnosis of superficial esophageal squamous cell carcinoma

作     者：Qian-Qian Meng Ye Gao Han Lin Tian-Jiao Wang Yan-Rong Zhang Jian Feng Zhao-Shen Li Lei Xin Luo-Wei Wang 

作者机构：Department of GastroenterologyChanghai HospitalShanghai 200433China Qingdao Medcare Digital Engineering Co.Ltd.Qingdao Medcare Digital Engineering Co.Ltd.Qingdao 26600Shandong ProvinceChina 

出 版 物：《World Journal of Gastroenterology》 (世界胃肠病学杂志（英文版）)

年 卷 期：2022年第28卷第37期

页      面：5483-5493页

核心收录：

学科分类：1002[医学-临床医学] 100214[医学-肿瘤学] 10[医学] 

基　　金：Supported by Shanghai Science and Technology Innovation Action Program, No. 21Y31900100 234 Clinical Research Fund of Changhai Hospital, No. 2019YXK006 

主　　题：Computer-aided diagnosis Artificial intelligence Deep learning Esophageal squamous cell carcinoma Early detection of cancer Upper gastrointestinal endoscopy 

摘      要：BACKGROUND Upper gastrointestinal endoscopy is critical for esophageal squamous cell carcinoma(ESCC)detection;however,endoscopists require long-term training to avoid missing superficial *** To develop a deep learning computer-assisted diagnosis(CAD)system for endoscopic detection of superficial ESCC and investigate its application *** We configured the CAD system for white-light and narrow-band imaging modes based on the YOLO v5 algorithm.A total of 4447 images from 837 patients and 1695 images from 323 patients were included in the training and testing datasets,*** experts and two non-expert endoscopists reviewed the testing dataset independently and with computer *** diagnostic performance was evaluated in terms of the area under the receiver operating characteristic curve,accuracy,sensitivity,and *** The area under the receiver operating characteristics curve,accuracy,sensitivity,and specificity of the CAD system were 0.982[95%confidence interval(CI):0.969-0.994],92.9%(95%CI:89.5%-95.2%),91.9%(95%CI:87.4%-94.9%),and 94.7%(95%CI:89.0%-97.6%),*** accuracy of CAD was significantly higher than that of non-expert endoscopists(78.3%,P0.001 compared with CAD)and comparable to that of expert endoscopists(91.0%,P=0.129 compared with CAD).After referring to the CAD results,the accuracy of the non-expert endoscopists significantly improved(88.2%vs 78.3%,P0.001).Lesions with Paris classification type 0-IIb were more likely to be inaccurately identified by the CAD *** The diagnostic performance of the CAD system is promising and may assist in improving detectability,particularly for inexperienced endoscopists.