{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T17:47:44Z","timestamp":1783619264853,"version":"3.55.0"},"reference-count":18,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,8,15]],"date-time":"2023-08-15T00:00:00Z","timestamp":1692057600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,8,15]],"date-time":"2023-08-15T00:00:00Z","timestamp":1692057600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100004826","name":"Natural Science Foundation of Beijing Municipality","doi-asserted-by":"publisher","award":["Z190024"],"award-info":[{"award-number":["Z190024"]}],"id":[{"id":"10.13039\/501100004826","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["11801301"],"award-info":[{"award-number":["11801301"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"CAMS Innovation Fund for Medical Sciences","award":["2020-I2M-C&T-B-005"],"award-info":[{"award-number":["2020-I2M-C&T-B-005"]}]},{"name":"Beijing Municipal Natural Science Foundation","award":["7212078"],"award-info":[{"award-number":["7212078"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Inform Decis Mak"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Differentiating between Crohn\u2019s disease (CD) and intestinal tuberculosis (ITB) with endoscopy is challenging. We aim to perform more accurate endoscopic diagnosis between CD and ITB by building a trustworthy AI differential diagnosis application.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      A total of 1271 electronic health record (EHR) patients who had undergone colonoscopies at Peking Union Medical College Hospital (PUMCH) and were clinically diagnosed with CD (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u2009875) or ITB (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u2009396) were used in this study. We build a workflow to make diagnoses with EHRs and mine differential diagnosis features; this involves finetuning the pretrained language models, distilling them into a light and efficient TextCNN model, interpreting the neural network and selecting differential attribution features, and then adopting manual feature checking and carrying out debias training.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The accuracy of debiased TextCNN on differential diagnosis between CD and ITB is 0.83 (CR F1: 0.87, ITB F1: 0.77), which is the best among the baselines. On the noisy validation set, its accuracy was 0.70 (CR F1: 0.87, ITB: 0.69), which was significantly higher than that of models without debias. We also find that the debiased model more easily mines the diagnostically significant features. The debiased TextCNN unearthed 39 diagnostic features in the form of phrases, 17 of which were key diagnostic features recognized by the guidelines.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>We build a trustworthy AI differential diagnosis application for differentiating between CD and ITB focusing on accuracy, interpretability and robustness. The classifiers perform well, and the features which had statistical significance were in agreement with clinical guidelines.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12911-023-02257-6","type":"journal-article","created":{"date-parts":[[2023,8,15]],"date-time":"2023-08-15T06:02:04Z","timestamp":1692079324000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":18,"title":["Building a trustworthy AI differential diagnosis application for Crohn\u2019s disease and intestinal tuberculosis"],"prefix":"10.1186","volume":"23","author":[{"given":"Keming","family":"Lu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuanren","family":"Tong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Si","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yucong","family":"Lin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yingyun","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hui","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6799-1812","authenticated-orcid":false,"given":"Yue","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sheng","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,8,15]]},"reference":[{"key":"2257_CR1","doi-asserted-by":"crossref","unstructured":"He Y, Zhu Z, Chen Y, Chen F, Wang Y, Ouyang C, ... Chen M. Development and validation of a novel diagnostic Nomogram to differentiate between intestinal tuberculosis and Crohn's disease: a 6-year prospective multicenter study. J Am College Gastroenterol. 2019;114(3):490\u2013499.","DOI":"10.14309\/ajg.0000000000000064"},{"key":"2257_CR2","first-page":"3319","volume-title":"International Conference on Machine Learning","author":"M Sundararajan","year":"2017","unstructured":"Sundararajan M, Taly A, Yan Q. Axiomatic attribution for deep networks. In: International Conference on Machine Learning. 2017. p. 3319\u201328 PMLR."},{"key":"2257_CR3","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1631","volume-title":"Incorporating priors with feature attribution on text classification","author":"F Liu","year":"2019","unstructured":"Liu F, Avci B. Incorporating priors with feature attribution on text classification. 2019. arXiv preprint arXiv:1906.08286."},{"key":"2257_CR4","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.494","volume-title":"Generating hierarchical explanations on text classification via feature interaction detection","author":"H Chen","year":"2020","unstructured":"Chen H, Zheng G, Ji Y. Generating hierarchical explanations on text classification via feature interaction detection. 2020. arXiv preprint arXiv:2004.02015."},{"key":"2257_CR5","volume-title":"Hierarchical interpretations for neural network predictions","author":"C Singh","year":"2018","unstructured":"Singh C, Murdoch WJ, Yu B. Hierarchical interpretations for neural network predictions. 2018. arXiv preprint arXiv:1806.05337."},{"key":"2257_CR6","doi-asserted-by":"publisher","first-page":"103865","DOI":"10.1016\/j.compbiomed.2020.103865","volume":"123","author":"M Graziani","year":"2020","unstructured":"Graziani M, Andrearczyk V, Marchand-Maillet S, M\u00fcller H. Concept attribution: Explaining cnn decisions tophysicians. Comput Biol Med. 2020;123:103865.","journal-title":"Comput Biol Med"},{"key":"2257_CR7","doi-asserted-by":"publisher","first-page":"118869","DOI":"10.1109\/ACCESS.2020.3005510","volume":"8","author":"S Hu","year":"2020","unstructured":"Hu S, Gao Y, Niu Z, Jiang Y, Li L, Xiao X, Wang M, Fang EF, Menpes-Smith W, Xia J, et al. Weakly supervised deep learning for covid-19 infection detection and classification from ct images. IEEE Access. 2020;8:118869\u201383.","journal-title":"IEEE Access."},{"key":"2257_CR8","doi-asserted-by":"crossref","unstructured":"Preuer K, Klambauer G, Rippmann F, et al. Interpretable deep learning in drug discovery[J]. Explainable AI: interpreting, explaining and visualizing deep learning. 2019. p. 331-45.","DOI":"10.1007\/978-3-030-28954-6_18"},{"issue":"1","key":"2257_CR9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-020-17431-x","volume":"11","author":"SM Lauritsen","year":"2020","unstructured":"Lauritsen SM, Kristensen M, Olsen MV, Larsen MS, Lauritsen KM, J\u00f8rgensen MJ, Lange J, Thiesson B. Explainable artificial intelligence model to predict acute critical illness from electronic health records. Nat Commun. 2020;11(1):1\u201311.","journal-title":"Nat Commun"},{"issue":"4","key":"2257_CR10","doi-asserted-by":"publisher","first-page":"552","DOI":"10.1016\/j.ophtha.2018.11.016","volume":"126","author":"R Sayres","year":"2019","unstructured":"Sayres R, Taly A, Rahimy E, Blumer K, Coz D, Hammel N, Krause J, Narayanaswamy A, Rastegar Z, Derek Wu, et al. Using a deep learning algorithm and integrated gradients explanation to assist grading for diabetic retinopathy. Ophthalmology. 2019;126(4):552\u201364.","journal-title":"Ophthalmology"},{"key":"2257_CR11","volume-title":"Pre-training with whole word masking for chinese bert","author":"Y Cui","year":"2019","unstructured":"Cui Y, Che W, Liu T, Qin B, Yang Z, Wang S, Hu G. Pre-training with whole word masking for chinese bert. 2019. arXiv preprint arXiv:1906.08101."},{"key":"2257_CR12","volume-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","author":"J Devlin","year":"2018","unstructured":"Devlin J, Chang MW, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. 2018. arXiv preprint arXiv:1810.04805."},{"key":"2257_CR13","volume-title":"Chinese lexical analysis with deep bi-gru-crf network","author":"J Zhenyu","year":"2018","unstructured":"Zhenyu J, Sun S, Sun K. Chinese lexical analysis with deep bi-gru-crf network. 2018. arXiv preprint arXiv:1807.01882."},{"key":"2257_CR14","first-page":"2980","volume-title":"Proceedings of the IEEE international conference on computer vision","author":"TY Lin","year":"2017","unstructured":"Lin TY, Goyal P, Girshick R, He K, Doll\u00e1r P. Focal loss for dense object detection. In: Proceedings of the IEEE international conference on computer vision. 2017. p. 2980\u20138."},{"key":"2257_CR15","volume-title":"Convolutional neural networks for sentence classification. corr abs\/1408.5882","author":"Y Kim","year":"2014","unstructured":"Kim Y. Convolutional neural networks for sentence classification. corr abs\/1408.5882. 2014. arXiv preprint arXiv:1408.5882."},{"key":"2257_CR16","volume-title":"Distilling the knowledge in a neural network","author":"G Hinton","year":"2015","unstructured":"Hinton G, Vinyals O, Dean J. Distilling the knowledge in a neural network. 2015. arXiv preprint arXiv:1503.02531."},{"key":"2257_CR17","doi-asserted-by":"crossref","unstructured":"Lee YJ, Yang SK, Byeon JS, Myung SJ, Chang HS, Hong SS, ... Yu CS. Analysis of colonoscopic findings in the differential diagnosis between intestinal tuberculosis and Crohn\u2019s disease.\u00a0Endoscopy. 2006;38(06):592\u2013597.","DOI":"10.1055\/s-2006-924996"},{"key":"2257_CR18","doi-asserted-by":"crossref","unstructured":"Tong Y, Lu K, Yang Y, Li J, Lin Y, Wu D, ... Qian J. Can natural language processing help differentiate inflammatory intestinal diseases in China? Models applying random forest and convolutional neural network approaches.\u00a0BMC Med Inform Decision Making. 2020;20(1):1\u20139.","DOI":"10.1186\/s12911-020-01277-w"}],"container-title":["BMC Medical Informatics and Decision Making"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-023-02257-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12911-023-02257-6\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-023-02257-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,17]],"date-time":"2023-11-17T14:14:53Z","timestamp":1700230493000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcmedinformdecismak.biomedcentral.com\/articles\/10.1186\/s12911-023-02257-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,15]]},"references-count":18,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["2257"],"URL":"https:\/\/doi.org\/10.1186\/s12911-023-02257-6","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-1625845\/v1","asserted-by":"object"}]},"ISSN":["1472-6947"],"issn-type":[{"value":"1472-6947","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,15]]},"assertion":[{"value":"5 May 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 August 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"15 August 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Research approval was obtained from Peking Union Medical College Hospital\u2019s Ethics Committee (approval no. S-K894). All the patients had given informed consent. The study adhered to the Declaration of Helsinki.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"None declared.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"160"}}