{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T04:59:56Z","timestamp":1785214796197,"version":"3.55.0"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T00:00:00Z","timestamp":1785196800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Mol. Biosci."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>This study aimed to develop, validate and evaluate interpretable machine learning models using clinical and laboratory data for prognosis prediction in patients with primary biliary cholangitis (PBC).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study included 7905 patients treated for PBC. The cohort data were randomly divided into training and testing sets, with external validation using an additional 2372 patients. Six machine learning models are compared in this research (logistic regression (LR), random forest (RF),Extreme Random Trees (ET), Extreme Gradient Boosting (XGBoost), Lightweight Gradient Boosting Machine (LightGBM), and Multi-Layer Perceptron (MLP)). Feature importance and model interpretation were analyzed using the SHapley Additive exPlanations (SHAP) method.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Ensemble tree models demonstrated significantly superior performance compared to traditional linear models and shallow neural networks in predicting PBC patient prognosis. Among these, the Extreme Random Tree model exhibited optimal predictive efficacy on both the training set (AUC = 0.9898) and external validation set (AUC = 0.9681), while the Random Forest model showed comparable performance with greater stability. At the patient level, SHAP\u2019s force maps and decision trees provided clinically meaningful explanations for the et algorithm. The bilirubin_albumin_ratio emerged as the core feature for predicting PBC prognosis, with bilirubin, n_days, and prothrombin serving as key influencing factors. The influence patterns of these features align closely with clinical and pathological mechanisms.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The ET model constructed in this study enables precise prognosis prediction for PBC patients. After SHAP analysis, it demonstrates good interpretability. The key prognostic features identified by the model provide quantitative evidence for clinically assessing disease severity in PBC patients and offer data support for developing individualized clinical intervention plans.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmolb.2026.1900864","type":"journal-article","created":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T04:25:09Z","timestamp":1785212709000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Development and evaluation of explainable machine learning models for predicting prognosis in patients with primary biliary cholangitis"],"prefix":"10.3389","volume":"13","author":[{"given":"Yifeng","family":"Dou","sequence":"first","affiliation":[{"name":"Network Information Center, Tianjin Medical University Baodi Hospital","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiantao","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Traditional Chinese Medicine, Tianjin Medical University Baodi Hospital","place":["Tianjin, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,28]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s12626-025-00190-w","article-title":"Boosting machine learning accuracy for cardiac disease prediction: the role of advanced feature engineering and model optimization","volume":"19","author":"Bilal","year":"2025","journal-title":"Rev. Socionetwork Strategies"},{"key":"B2","volume-title":"Variable Importance in Recursive Feature Selection with Random Forests for Mortality Prediction in ICU","author":"Boughorbel","year":"2015"},{"key":"B3","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1186\/s12916-019-1466-7","article-title":"Calibration: the achilles' heel of predictive analytics and On behalf of Topic Group \u2018Evaluating diagnostic tests and prediction models' of the STRATOS initiative","volume":"17","author":"Calster","year":"2019","journal-title":"BMC Med."},{"key":"B4","doi-asserted-by":"crossref","DOI":"10.1145\/2939672.2939785","volume-title":"Xgboost: Extreme Gradient Boosting","author":"Chen","year":"2016"},{"key":"B5","doi-asserted-by":"publisher","first-page":"946","DOI":"10.1159\/000531557","article-title":"The albumin-bilirubin score as a predictor of liver-related mortality in primary biliary cholangitis with compensated cirrhosis","volume":"41","author":"Chen","year":"2023","journal-title":"Dig. Dis."},{"key":"B6","doi-asserted-by":"publisher","first-page":"871","DOI":"10.1002\/hep.22428","article-title":"Biochemical response to ursodeoxycholic acid and long-term prognosis in primary biliary cirrhosis","volume":"48","author":"Corpechot","year":"2008","journal-title":"Hepatology"},{"key":"B7","volume-title":"Unsupervised Extremely Randomized Trees[C]\/\/Pacific-asia Conference on Knowledge Discovery and Data Mining","author":"Dalleau","year":"2018"},{"key":"B8","doi-asserted-by":"publisher","first-page":"145","DOI":"10.1016\/j.jhep.2017.03.022","article-title":"EASL clinical practice guidelines: the diagnosis and management of patients with primary biliary cholangitis","volume":"67","year":"2017","journal-title":"J. Hepatol."},{"key":"B9","doi-asserted-by":"publisher","first-page":"61","DOI":"10.7150\/ijms.88481","article-title":"Developing a prognostic model for primary biliary cholangitis based on a random survival forest model","volume":"21","author":"Fu","year":"2024","journal-title":"Int. Journal Medical Sciences"},{"key":"B10","doi-asserted-by":"publisher","first-page":"175","DOI":"10.1007\/s12016-019-08772-7","article-title":"Clinical management of primary biliary Cholangitis- strategies and evolving trends","volume":"59","author":"Gao","year":"2020","journal-title":"Clin. Rev. Allergy and Immunol."},{"key":"B11","doi-asserted-by":"publisher","first-page":"101198","DOI":"10.1016\/j.jhepr.2024.101198","article-title":"Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis","volume":"7","author":"Gerussi","year":"2025","journal-title":"JHEP Rep."},{"key":"B12","doi-asserted-by":"publisher","first-page":"2647","DOI":"10.1111\/j.1572-0241.2002.06043.x","article-title":"Long-term ursodiol treatment outcomes for patients with primary biliary cirrhosis","volume":"97","author":"Jorgensen","year":"2002","journal-title":"Am. J. Gastroenterology"},{"key":"B13","unstructured":"2023"},{"key":"B14","doi-asserted-by":"publisher","first-page":"S180","DOI":"10.4103\/ijcm.ijcm_454_24","article-title":"Fivefold cross-validation approach in evaluating the robustness of machine learning models for prediction of esophageal cancer","volume":"51","author":"Kalita","year":"2026","journal-title":"Indian J. Community Med."},{"key":"B15","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1007\/978-3-031-27700-9_6","article-title":"Logistic regression-based machine learning model for mutation classification in the discovery of precision medicine","author":"Kathiresan","year":"2023","journal-title":"EAI\/Springer Innovations Commun. Comput."},{"key":"B16","doi-asserted-by":"publisher","first-page":"183","DOI":"10.4018\/979-8-3693-1243-8.ch010","article-title":"Enhanced random forest approach in digital healthcare in medicine recommendation and classification","author":"Kaur","year":"2024","journal-title":"Adv. Med. Diagnosis, Treat. Care"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1111\/hepr.13966","article-title":"Development and validation of machine learning model for predicting treatment responders in patients with primary biliary cholangitis","volume":"54","author":"Kimura","year":"2024","journal-title":"Hepatology Research The Official Journal Jpn. Soc. Hepatology"},{"key":"B18","doi-asserted-by":"publisher","first-page":"034","DOI":"10.1055\/s-0039-1697617","article-title":"The pathogenesis of primary biliary cholangitis: a comprehensive review","volume":"40","author":"Lleo","year":"2020","journal-title":"Seminars Liver Dis."},{"key":"B19","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1038\/s42256-019-0138-9","article-title":"From local explanations to global understanding with explainable AI for trees","volume":"2","author":"Lundberg","year":"2020","journal-title":"Nat. Mach. Intell."},{"key":"B20","volume-title":"Lightgbm: A Highly Efficient Gradient Boosting Decision Tree[C]\/\/Neural Information Processing Systems","author":"Meng","year":"2017"},{"key":"B21","volume-title":"Decision Curve Analysis Explained","author":"Montero","year":"2026"},{"key":"B22","doi-asserted-by":"publisher","first-page":"721","DOI":"10.1007\/s43441-024-00640-7","article-title":"Improvement of midpoint imputation for estimation of median survival time for interval-censored time-to-event data","volume":"58","author":"Nakagawa","year":"2024","journal-title":"Ther. Innovation and Regulatory Science"},{"key":"B23","doi-asserted-by":"publisher","first-page":"e70566","DOI":"10.1002\/eng2.70566","article-title":"Impact of feature engineering on machine learning models for heart disease prediction across multiple data sources","volume":"8","author":"Shil","year":"2026","journal-title":"Eng. Rep."},{"key":"B24","doi-asserted-by":"publisher","first-page":"187","DOI":"10.14218\/JCTH.2020.00127","article-title":"Predictive model of ursodeoxycholic acid treatment response in primary biliary cholangitis","volume":"9","author":"Shu","year":"2021","journal-title":"J. Clinical Translational Hepatology"},{"key":"B25","doi-asserted-by":"publisher","first-page":"2018","DOI":"10.1109\/78.215329","article-title":"Image restoration using a multilayer perceptron with a multilevel sigmoidal function","volume":"41","author":"Sivakumar","year":"2002","journal-title":"IEEE Trans. Signal Process."},{"key":"B26","doi-asserted-by":"publisher","first-page":"63","DOI":"10.2147\/HMER.S361077","article-title":"Primary biliary cholangitis: promising emerging innovative therapies and their impact on GLOBE scores","volume":"15","author":"Sohal","year":"2023","journal-title":"Hepat. Med."},{"key":"B27","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1007\/s13349-024-00776-y","article-title":"Missing data imputation model for dam health monitoring based on mode decomposition and deep learning","volume":"14","author":"Song","year":"2024","journal-title":"J. Civ. Struct. Health Monit."},{"key":"B28","doi-asserted-by":"publisher","first-page":"2062","DOI":"10.1016\/j.cgh.2024.05.008","article-title":"Development and validation of a scoring system to predict response to obeticholic acid in primary biliary cholangitis","volume":"22","author":"Vincentis","year":"2024","journal-title":"Clin. Gastroenterology Hepatology"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1426","DOI":"10.12449\/JCH250730","article-title":"Research advances in prognostic models for primary biliary cholangitis","volume":"41","author":"Wan","year":"2025","journal-title":"J. Clin. Hepatol."},{"key":"B30","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s12072-021-10276-6","article-title":"APASL clinical practice guidance: the diagnosis and management of patients with primary biliary cholangitis","volume":"16","author":"You","year":"2022","journal-title":"Hepatol. Int."},{"key":"B31","doi-asserted-by":"publisher","first-page":"2541","DOI":"10.1016\/j.jss.2012.05.073","article-title":"Nearest neighbor selection for iteratively kNN imputation","volume":"85","author":"Zhang","year":"2012","journal-title":"J. Syst. and Softw."},{"key":"B32","doi-asserted-by":"publisher","first-page":"1268","DOI":"10.1007\/s00330-023-10080-w","article-title":"Noninvasive prediction of insufficient biochemical response after ursodeoxycholic acid treatment in patients with primary biliary cholangitis based on pretreatment nonenhanced MRI","volume":"34","author":"Zhang","year":"2024","journal-title":"Eur. Radiology"},{"key":"B33","doi-asserted-by":"publisher","first-page":"e3639","DOI":"10.1002\/brb3.3639","article-title":"Depressive symptom as a risk factor for cirrhosis in patients with primary biliary cholangitis: analysis based on Lasso-logistic regression and decision tree models","volume":"14","author":"Zhou","year":"2024","journal-title":"Behavior"},{"key":"B34","doi-asserted-by":"publisher","first-page":"31369","DOI":"10.1038\/s41598-024-82854-1","article-title":"Development and external validation of an early prediction model to identify non-responsive patients and prognosis of UDCA treatment in primary biliary cholangitis","volume":"14","author":"Zhu","year":"2024","journal-title":"Sci. Rep."}],"container-title":["Frontiers in Molecular Biosciences"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmolb.2026.1900864\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T04:25:12Z","timestamp":1785212712000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmolb.2026.1900864\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,28]]},"references-count":34,"alternative-id":["10.3389\/fmolb.2026.1900864"],"URL":"https:\/\/doi.org\/10.3389\/fmolb.2026.1900864","relation":{},"ISSN":["2296-889X"],"issn-type":[{"value":"2296-889X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,28]]},"article-number":"1900864"}}