Xun Li, Chenxia Zhang, L. Yao, Jun Zhang, Kun Zhang, Hui Feng, H. Yu
{"title":"A deep learning–based system to identify originating mural layer of upper gastrointestinal submucosal tumors under EUS","authors":"Xun Li, Chenxia Zhang, L. Yao, Jun Zhang, Kun Zhang, Hui Feng, H. Yu","doi":"10.1097/eus.0000000000000029","DOIUrl":null,"url":null,"abstract":"EUS is the most accurate procedure to determine the originating mural layer and subsequently select the treatment of submucosal tumors (SMTs). However, it requires superb technical and cognitive skills. In this study, we propose a system named SMT Master to determine the originating mural layer of SMTs under EUS. We developed 3 models: deep convolutional neural network (DCNN) 1 for lesion segmentation, DCNN2 for mural layer segmentation, and DCNN3 for the originating mural layer classification. A total of 2721 EUS images from 201 patients were used to train the 3 models. We validated our model internally and externally using 283 images from 26 patients and 172 images from 26 patients, respectively. We applied 368 images from 30 patients for the man-machine contest and used 30 video clips to test the originating mural layer classification. In the originating mural layer classification task, DCNN3 achieved a classification accuracy of 84.43% and 80.68% at internal and external validations, respectively. In the video test, the accuracy was 80.00%. DCNN1 achieved Dice coefficients of 0.956 and 0.776 for lesion segmentation at internal and external validations, respectively, whereas DCNN2 achieved Dice coefficients of 0.820 and 0.740 at internal and external validations, respectively. The system achieved 90.00% accuracy in classification, which is comparable with that of EUS experts. Our proposed system has the potential to solve difficulties in determining the originating mural layer of SMTs in EUS procedures, which relieves the EUS learning pressure of physicians.","PeriodicalId":11577,"journal":{"name":"Endoscopic Ultrasound","volume":"253 ","pages":""},"PeriodicalIF":4.4000,"publicationDate":"2023-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Endoscopic Ultrasound","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/eus.0000000000000029","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GASTROENTEROLOGY & HEPATOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
EUS is the most accurate procedure to determine the originating mural layer and subsequently select the treatment of submucosal tumors (SMTs). However, it requires superb technical and cognitive skills. In this study, we propose a system named SMT Master to determine the originating mural layer of SMTs under EUS. We developed 3 models: deep convolutional neural network (DCNN) 1 for lesion segmentation, DCNN2 for mural layer segmentation, and DCNN3 for the originating mural layer classification. A total of 2721 EUS images from 201 patients were used to train the 3 models. We validated our model internally and externally using 283 images from 26 patients and 172 images from 26 patients, respectively. We applied 368 images from 30 patients for the man-machine contest and used 30 video clips to test the originating mural layer classification. In the originating mural layer classification task, DCNN3 achieved a classification accuracy of 84.43% and 80.68% at internal and external validations, respectively. In the video test, the accuracy was 80.00%. DCNN1 achieved Dice coefficients of 0.956 and 0.776 for lesion segmentation at internal and external validations, respectively, whereas DCNN2 achieved Dice coefficients of 0.820 and 0.740 at internal and external validations, respectively. The system achieved 90.00% accuracy in classification, which is comparable with that of EUS experts. Our proposed system has the potential to solve difficulties in determining the originating mural layer of SMTs in EUS procedures, which relieves the EUS learning pressure of physicians.
期刊介绍:
Endoscopic Ultrasound, a publication of Euro-EUS Scientific Committee, Asia-Pacific EUS Task Force and Latin American Chapter of EUS, is a peer-reviewed online journal with Quarterly print on demand compilation of issues published. The journal’s full text is available online at http://www.eusjournal.com. The journal allows free access (Open Access) to its contents and permits authors to self-archive final accepted version of the articles on any OAI-compliant institutional / subject-based repository. The journal does not charge for submission, processing or publication of manuscripts and even for color reproduction of photographs.