Artificial intelligence-based diagnosis of standard endoscopic ultrasonography scanning sites in the biliopancreatic system: a multicenter retrospective study

作者全名："Tian, Shuxin; Shi, Huiying; Chen, Weigang; Li, Shijie; Han, Chaoqun; Du, Fan; Wang, Weijun; Wen, Hongxu; Lei, Yali; Deng, Liang; Tang, Jing; Zhang, Jinjie; Lin, Jianjiao; Shi, Lei; Ning, Bo; Zhao, Kui; Miao, Jiarong; Wang, Guobao; Hou, Hui; Huang, Xiaoxi; Kong, Wenjie; Jin, Xiaojuan; Ding, Zhen; Lin, Rong"

作者地址："[Tian, Shuxin; Shi, Huiying; Han, Chaoqun; Du, Fan; Wang, Weijun; Ding, Zhen; Lin, Rong] Huazhong Univ Sci & Technol, Union Hosp, Tongji Med Coll, Dept Gastroenterol, Wuhan 430022, Peoples R China; [Tian, Shuxin; Chen, Weigang] Shihezi Univ, Affiliated Hosp 1, Med Coll, Dept Gastroenterol, Shihezi, Peoples R China; [Tian, Shuxin; Chen, Weigang; Li, Shijie] Natl Hlth Commiss, Key Lab Cent Asia High Incidence Dis Prevent & Con, Shihezi, Peoples R China; [Li, Shijie] Peking Univ Canc Hosp & Inst, Minist Educ Beijing, Dept Endoscopy Ctr, Key Lab Carcinogenesis & Translat Res, Beijing, Peoples R China; [Wen, Hongxu] Lanzhou Second Peoples Hosp, Dept Gastroenterol, Lanzhou, Peoples R China; [Lei, Yali] Weinan Cent Hosp, Dept Gastroenterol, Weinan, Peoples R China; [Deng, Liang] Chongqing Med Univ, Affiliated Hosp 1, Dept Gastroenterol, Chongqing, Peoples R China; [Tang, Jing] Chongqing Univ, Fuling Hosp, Dept Gastroenterol, Chongqing, Peoples R China; [Ning, Bo] Chongqing Med Univ, Affiliated Hosp 2, Dept Gastroenterol, Chongqing, Peoples R China; [Lin, Jianjiao] Longgang Dist Peoples Hosp, Dept Gastroenterol, Shenzhen, Peoples R China; [Shi, Lei] Southwest Med Univ, Affiliated Hosp, Dept Gastroenterol, Luzhou, Peoples R China; [Zhang, Jinjie] Baotou Med Coll, Affiliated Hosp 2, Dept Gastroenterol, Baotou, Peoples R China; [Zhao, Kui] Chendu Med Coll, Affiliated Hosp 1, Dept Gastroenterol, Chengdu, Peoples R China; [Miao, Jiarong] Kunming Med Univ, Affiliated Hosp 1, Dept Gastroenterol, Kunming, Peoples R China; [Miao, Jiarong] Yunnan Prov Clin Res Ctr Digest Dis, Kunming, Peoples R China; [Wang, Guobao] Sun Yat Aen Univ, Dept endoscopy, Canc Ctr, Guangzhou, Peoples R China; [Ding, Zhen] Sun Yat Sen Univ, Affiliated Hosp 1, Dept Endoscopy Ctr, Guangzhou, Peoples R China; [Huang, Xiaoxi] Haikou Peoples Hosp, Dept Gastroenterol, Haikou, Peoples R China; [Kong, Wenjie] Peoples Hosp Xinjiang Autonomous Reg, Dept Gastroenterol, Urumqi, Peoples R China; [Hou, Hui] Xinjiang Med Univ, Affiliated Hosp 5, Dept Gastroenterol, Urumqi, Peoples R China; [Jin, Xiaojuan] Suining Cent Hosp, Dept Gastroenterol, Suining, Peoples R China"

通信作者："Ding, Z; Lin, R (通讯作者)，Huazhong Univ Sci & Technol, Union Hosp, Tongji Med Coll, Dept Gastroenterol, Wuhan 430022, Peoples R China."

来源：INTERNATIONAL JOURNAL OF SURGERY

ESI学科分类：CLINICAL MEDICINE

WOS号：WOS:001185065100013

JCR分区：Q1

影响因子：15.3

年份：2024

卷号：110

期号：3

开始页：1637

结束页：1644

文献类型：Article

关键词：artificial intelligence; biliary disease; endoscopic ultrasonography; imaging; pancreatic disease

摘要："Background:There are challenges for beginners to identify standard biliopancreatic system anatomical sites on endoscopic ultrasonography (EUS) images. Therefore, the authors aimed to develop a convolutional neural network (CNN)-based model to identify standard biliopancreatic system anatomical sites on EUS images.Methods:The standard anatomical structures of the gastric and duodenal regions observed by EUS was divided into 14 sites. The authors used 6230 EUS images with standard anatomical sites selected from 1812 patients to train the CNN model, and then tested its diagnostic performance both in internal and external validations. Internal validation set tests were performed on 1569 EUS images of 47 patients from two centers. Externally validated datasets were retrospectively collected from 16 centers, and finally 131 patients with 85 322 EUS images were included. In the external validation, all EUS images were read by CNN model, beginners, and experts, respectively. The final decision made by the experts was considered as the gold standard, and the diagnostic performance between CNN model and beginners were compared.Results:In the internal test cohort, the accuracy of CNN model was 92.1-100.0% for 14 standard anatomical sites. In the external test cohort, the sensitivity and specificity of CNN model were 89.45-99.92% and 93.35-99.79%, respectively. Compared with beginners, CNN model had higher sensitivity and specificity for 11 sites, and was in good agreement with the experts (Kappa values 0.84-0.98).Conclusions:The authors developed a CNN-based model to automatically identify standard anatomical sites on EUS images with excellent diagnostic performance, which may serve as a potentially powerful auxiliary tool in future clinical practice."

基金机构："National Natural Science Foundation of China [81974068, 82170571]; National key research and development program of China [2017YFC0 110003]; Natural Science Foundation of Hubei Province [2017CFA061]; Non-profit Central Research Institute Fund of Chinese Academy of Medical Sciences [2020- PT330-003]"

基金资助正文："Supported by the National Natural Science Foundation of China (Nos. 81974068 and 82170571), the National key research and development program of China (No. 2017YFC0 110003), the Natural Science Foundation of Hubei Province (No. 2017CFA061), the Non-profit Central Research Institute Fund of Chinese Academy of Medical Sciences (No. 2020- PT330-003). The funding body had no part in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript."