{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T14:53:30Z","timestamp":1791903210265,"version":"4.3.4"},"reference-count":22,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"\n Background<\/jats:title>\n Small bowel capsule endoscopy (SBCE) is widely used to evaluate small bowel involvement in Crohn disease (CD); however, manual Lewis score calculation is time-consuming and susceptible to interobserver variability.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop and preliminarily validate a multitask deep learning system for automated Lewis score estimation in patients with small bowel CD undergoing SBCE.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We developed Eval-Model, a multitask framework integrating You Only Look Once (YOLO)\u2013style lesion detection and localization for ulcers and edema, a stenosis classification branch, and a long short-term memory\u2013based temporal regression module. The primary dataset comprised SBCE examinations from 814 patients at the Children\u2019s Hospital of Fudan University. In this dataset, 725 (89.1%) cases were included for lesion-level annotation, and 89 (10.9%) cases with expert-consensus Lewis scores were used for scoring model development and internal testing. These 89 cases comprised 49 (55.1%) training cases and 40 (44.9%) internal test cases. All splits were performed at the patient level. A retrospective same-center temporal validation cohort of 20 consecutive cases from 2026 and 5 control cases were used for preliminary assessments of temporal robustness and specificity.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n In the 40-case internal Lewis score test set, Eval-Model achieved a mean squared error of 14.693 (95% CI 8.72-22.15), root mean squared error of 3.833 (95% CI 2.95-4.71), mean absolute error of 2.9 (95% CI 2.1-3.8), R2 of 0.82 (95% CI 0.70-0.90), and explained variance of 0.83 (95% CI 0.72-0.91). For lesion recognition, the model achieved an overall mean average precision of 0.94 with an inference time of 15 ms per frame. In the 20-case same-center temporal validation cohort, the mean squared error was 26.848 (95% CI 12.51-45.33), the root mean squared error was 5.182 (95% CI 3.54-6.73), the mean absolute error was 3.3 (95% CI 2.2-4.6), R2 rounded to 1.00 (95% CI 0.98-1.00), and the explained variance rounded to 1.00 (95% CI 0.98-1.00). These temporal estimates require cautious interpretation because of the small, same-center cohort.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Eval-Model demonstrated preliminary feasibility for automated Lewis score estimation from SBCE in small bowel CD. The system may serve as a standardized adjunct to expert review, but larger multicenter prospective validation with independent external annotation is required before clinical deployment.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/92416","type":"journal-article","created":{"date-parts":[[2026,9,9]],"date-time":"2026-09-09T16:25:05Z","timestamp":1788971105000},"page":"e92416","source":"Crossref","is-referenced-by-count":0,"title":["Automated Lewis Score Estimation for Small Bowel Crohn Disease Using Capsule Endoscopy: Retrospective Cohort Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9542-7933","authenticated-orcid":false,"given":"Yingying","family":"Meng","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4969-2488","authenticated-orcid":false,"given":"Min","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6831-3981","authenticated-orcid":false,"given":"Jie","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4931-4006","authenticated-orcid":false,"given":"Ying","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,10,13]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(16)31711-1"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1159\/000527352"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab5086"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1111\/j.1365-2036.2007.03556.x"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1016\/j.gie.2020.01.054"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1053\/j.gastro.2019.06.025"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1155\/2019\/7546215"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.3390\/s19061265"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2022.01.135"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1016\/j.physd.2019.132306"},{"key":"ref11","author":"Girshick, R","year":"2015","journal-title":"2015 IEEE International Conference on Computer Vision (ICCV)"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1712.03452"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1093\/ecco-jcc\/jjab117"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1093\/ecco-jcc\/jjaa234"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.cgh.2025.09.036"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1055\/a-1881-4209"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1111\/den.14670"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1093\/ecco-jcc\/jjv166"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1055\/s-0034-1390894"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1016\/s2468-1253(19)30088-3"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1016\/j.gie.2020.05.066"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1177\/17562848241251569"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e92416","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T14:35:07Z","timestamp":1791902107000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e92416"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,13]]},"references-count":22,"URL":"https:\/\/doi.org\/10.2196\/92416","relation":{"has-preprint":[{"id-type":"doi","id":"10.2196\/preprints.92416","asserted-by":"object"}]},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,13]]}}}