{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T06:35:51Z","timestamp":1778826951616,"version":"3.51.4"},"reference-count":35,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,5,12]],"date-time":"2025-05-12T00:00:00Z","timestamp":1747008000000},"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. Digit. Health"],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Heart Failure (HF) complicated by thyroid dysfunction presents a complex clinical challenge, demanding more advanced risk stratification tools. In this study, we propose an AI-driven machine learning (ML) approach to predict mortality and hospitalization risk in HF patients with coexisting thyroid disorders.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Using a retrospective cohort of 762 HF patients (including euthyroid, hypothyroid, hyperthyroid, and low T3 syndrome cases), we developed and optimized several ML models\u2014including Random Forest, Gradient Boosting, Support Vector Machines, and others\u2014to identify high-risk individuals.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The best-performing model, a Random Forest classifier, achieved robust predictive accuracy for both 1-year mortality and HF-related hospitalization (area under the ROC curve \u223c0.80 for each). We further employed model interpretability techniques (Local Interpretable Model-agnostic Explanations, LIME) to elucidate key predictors of risk at the individual level. This interpretability revealed that factors such as atrial fibrillation, absence of cardiac resynchronization therapy, amiodarone use, and abnormal thyroid-stimulating hormone (TSH) levels strongly influenced model predictions, providing clinicians with transparent insights into each prediction. Additionally, a multi-objective risk stratification analysis across thyroid status subgroups highlighted that patients with hypothyroidism and low T3 syndrome are particularly vulnerable under high-risk conditions, indicating a need for closer monitoring and tailored interventions in these groups.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>In summary, our study demonstrates an innovative AI methodology for medical risk prediction: interpretable ML models can accurately stratify mortality and hospitalization risk in HF patients with thyroid dysfunction, offering a novel tool for personalized medicine. These findings suggest that integrating explainable AI into clinical workflows can improve prognostic precision and inform targeted management, though prospective validation is warranted to confirm realworld applicability.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fdgth.2025.1583399","type":"journal-article","created":{"date-parts":[[2025,5,12]],"date-time":"2025-05-12T05:39:37Z","timestamp":1747028377000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":12,"title":["Interpretable AI-driven multi-objective risk prediction in heart failure patients with thyroid dysfunction"],"prefix":"10.3389","volume":"7","author":[{"given":"Massimo","family":"Iacoviello","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vito","family":"Santamato","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alessandro","family":"Pagano","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Agostino","family":"Marengo","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,5,12]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1667","DOI":"10.1007\/s11739-024-03665-w","article-title":"Subclinical hypothyroidism predicts outcome in heart failure: insights from the T.O.S.CA. Registry","volume":"19","author":"De Luca","year":"2024","journal-title":"Intern Emerg Med"},{"key":"B2","doi-asserted-by":"publisher","first-page":"743","DOI":"10.1016\/S2213-8587(23)00227-9","article-title":"The optimal healthy ranges of thyroid function defined by the risk of cardiovascular disease and mortality: systematic review and individual participant data meta-analysis","volume":"11","author":"Xu","year":"2023","journal-title":"Lancet Diabetes Endocrinol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"83","DOI":"10.1186\/s12872-019-1055-x","article-title":"Subclinical thyroid dysfunction is associated with adverse prognosis in heart failure patients with reduced ejection fraction","volume":"19","author":"Yang","year":"2019","journal-title":"BMC Cardiovasc Disord"},{"key":"B4","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1186\/s12955-023-02109-x","article-title":"Machine learning prognosis model based on patient-reported outcomes for chronic heart failure patients after discharge","volume":"21","author":"Tian","year":"2023","journal-title":"Health Qual Life Outcomes"},{"key":"B5","doi-asserted-by":"publisher","first-page":"109330","DOI":"10.1016\/j.compbiomed.2024.109330","article-title":"An efficient cardiovascular disease prediction model through AI-driven IoT technology","volume":"183","author":"Marengo","year":"2024","journal-title":"Comput Biol Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"2947","DOI":"10.3390\/diagnostics12122947","article-title":"Risk of mortality prediction involving time-varying covariates for patients with heart failure using deep learning","volume":"12","author":"Nakamura","year":"2022","journal-title":"Diagnostics"},{"key":"B7","doi-asserted-by":"publisher","first-page":"776","DOI":"10.3390\/life12060776","article-title":"Exploring and identifying prognostic phenotypes of patients with heart failure guided by explainable machine learning","volume":"12","author":"Zhou","year":"2022","journal-title":"Life"},{"key":"B8","doi-asserted-by":"publisher","first-page":"86","DOI":"10.2174\/187153012799278947","article-title":"Incidence and prevalence of hypothyroidism in patients affected by chronic heart failure: role of amiodarone","volume":"12","author":"Triggiani","year":"2012","journal-title":"Endocr Metab Immune Disord Drug Targets"},{"key":"B9","doi-asserted-by":"publisher","first-page":"437","DOI":"10.2174\/1871530319666191018134524","article-title":"Thyroid disorders and prognosis in chronic heart failure: a long-term follow-up study","volume":"20","author":"Iacoviello","year":"2020","journal-title":"Endocr Metab Immune Disord Drug Targets"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1935","DOI":"10.2174\/1871530321666210430131510","article-title":"TSH Variations in chronic heart failure outpatients: clinical correlates and outcomes","volume":"21","author":"Terlizzese","year":"2021","journal-title":"Endocr Metab Immune Disord Drug Targets"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1791","DOI":"10.1002\/ehf2.12742","article-title":"Clinical predictors of all-cause mortality in patients presenting to specialist heart failure clinic with raised NT-proBNP and no heart failure","volume":"7","author":"Garg","year":"2020","journal-title":"ESC Heart Fail"},{"key":"B12","doi-asserted-by":"publisher","first-page":"76","DOI":"10.1186\/s40001-024-01675-0","article-title":"Predictive modeling for acute kidney injury after percutaneous coronary intervention in patients with acute coronary syndrome: a machine learning approach","volume":"29","author":"Behnoush","year":"2024","journal-title":"Eur J Med Res"},{"key":"B13","doi-asserted-by":"publisher","first-page":"e0276835","DOI":"10.1371\/journal.pone.0276835","article-title":"A deep learning system for heart failure mortality prediction","volume":"18","author":"Li","year":"2023","journal-title":"PLoS One"},{"key":"B14","doi-asserted-by":"publisher","first-page":"ehad655.1102","DOI":"10.1093\/eurheartj\/ehad655.1102","article-title":"Machine learning to predict in-hospital outcomes in patients with acute heart failure","volume":"44","author":"Sibilia","year":"2023","journal-title":"Eur Heart J"},{"key":"B15","doi-asserted-by":"publisher","first-page":"103496","DOI":"10.1016\/j.autrev.2023.103496","article-title":"Machine learning application in autoimmune diseases: state of art and future prospectives","volume":"23","author":"Danieli","year":"2024","journal-title":"Autoimmun Rev"},{"key":"B16","doi-asserted-by":"publisher","first-page":"e19739","DOI":"10.2196\/19739","article-title":"An application of machine learning to etiological diagnosis of secondary hypertension: retrospective study using electronic medical records","volume":"9","author":"Diao","year":"2021","journal-title":"JMIR Med Inform"},{"key":"B17","doi-asserted-by":"publisher","first-page":"17313","DOI":"10.1038\/s41598-022-20176-w","article-title":"Identifying pre-existing conditions and multimorbidity patterns associated with in-hospital mortality in patients with COVID-19","volume":"12","author":"Bucholc","year":"2022","journal-title":"Sci Rep"},{"key":"B18","doi-asserted-by":"publisher","first-page":"150","DOI":"10.1186\/s12902-021-00809-3","article-title":"Prediction model for thyrotoxic atrial fibrillation: a retrospective study","volume":"21","author":"Ponomartseva","year":"2021","journal-title":"BMC Endocr Disord"},{"key":"B19","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1016\/j.jchf.2019.06.013","article-title":"Machine learning prediction of mortality and hospitalization in heart failure with preserved ejection fraction","volume":"8","author":"Angraal","year":"2020","journal-title":"JACC Heart Fail"},{"key":"B20","doi-asserted-by":"publisher","first-page":"e1920745","DOI":"10.1001\/jamanetworkopen.2019.20745","article-title":"Association of subclinical hypothyroidism and cardiovascular disease with mortality","volume":"3","author":"Inoue","year":"2020","journal-title":"JAMA Netw Open"},{"key":"B21","doi-asserted-by":"publisher","first-page":"844","DOI":"10.1001\/jamacardio.2022.1900","article-title":"Machine learning\u2013based models incorporating social determinants of health vs traditional models for predicting in-hospital mortality in patients with heart failure","volume":"7","author":"Segar","year":"2022","journal-title":"JAMA Cardiol"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1100","DOI":"10.1515\/med-2022-0508","article-title":"Risk prediction of cardiovascular disease using machine learning classifiers","volume":"17","author":"Pal","year":"2022","journal-title":"Open Med"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1018","DOI":"10.1016\/j.jvir.2019.11.030","article-title":"Machine learning offers exciting potential for predicting postprocedural outcomes: a framework for developing random forest models in IR","author":"Sinha","year":"2020","journal-title":"J Vasc Interv Radiol"},{"key":"B24","doi-asserted-by":"crossref","first-page":"505","DOI":"10.1007\/978-3-031-37111-0_35","article-title":"Assessment of public health performance in relation to hospital energy demand, socio-economic efficiency and quality of services: an Italian case study","volume-title":"Computational Science and Its Applications\u2014ICCSA 2023 Workshops","author":"Santamato","year":"2023"},{"key":"B25","doi-asserted-by":"publisher","first-page":"10144","DOI":"10.3390\/app142210144","article-title":"Exploring the impact of artificial intelligence on healthcare management: a combined systematic review and machine-learning approach","volume":"14","author":"Santamato","year":"2024","journal-title":"Appl Sci"},{"key":"B26","doi-asserted-by":"publisher","first-page":"6016","DOI":"10.3390\/app14146016","article-title":"Machine learning for evaluating hospital mobility: an Italian case study","volume":"14","author":"Santamato","year":"2024","journal-title":"Appl Sci"},{"key":"B27","doi-asserted-by":"publisher","first-page":"124020","DOI":"10.1016\/j.eswa.2024.124020","article-title":"Healthcare performance analytics based on the novel PDA methodology for assessment of efficiency and perceived quality outcomes: a machine learning approach","volume":"252","author":"Santamato","year":"2024","journal-title":"Expert Syst Appl"},{"key":"B28","doi-asserted-by":"publisher","first-page":"837","DOI":"10.3390\/app14020837","article-title":"Machine learning for predicting neurodevelopmental disorders in children","volume":"14","author":"Toki","year":"2024","journal-title":"Appl Sci"},{"key":"B29","doi-asserted-by":"publisher","first-page":"2657","DOI":"10.1016\/j.jacc.2017.03.571","article-title":"Artificial intelligence in precision cardiovascular medicine","volume":"69","author":"Krittanawong","year":"2017","journal-title":"J Am Coll Cardiol"},{"key":"B30","doi-asserted-by":"publisher","first-page":"604","DOI":"10.7326\/0003-4819-150-9-200905050-00006","article-title":"A new equation to estimate glomerular filtration rate","volume":"150","author":"Levey","year":"2009","journal-title":"Ann Intern Med"},{"key":"B31","doi-asserted-by":"publisher","first-page":"e24374","DOI":"10.2196\/24374","article-title":"Shaping workflows in digital and remote diabetes care during the COVID-19 pandemic via service design: prospective, longitudinal, open-label feasibility trial","volume":"9","author":"Braune","year":"2021","journal-title":"JMIR Mhealth Uhealth"},{"key":"B32","doi-asserted-by":"publisher","first-page":"103627","DOI":"10.1016\/j.jbi.2020.103627","article-title":"A review on deep learning approaches in healthcare systems: taxonomies, challenges, and open issues","volume":"113","author":"Shamshirband","year":"2021","journal-title":"J Biomed Inform"},{"key":"B33","doi-asserted-by":"publisher","first-page":"295","DOI":"10.1186\/s12911-020-01318-4","article-title":"Predicting mortality in critically ill patients with diabetes using machine learning and clinical notes","volume":"20","author":"Ye","year":"2020","journal-title":"BMC Med Inform Decis Mak"},{"key":"B34","doi-asserted-by":"publisher","first-page":"e0155705","DOI":"10.1371\/journal.pone.0155705","article-title":"Application of machine learning techniques to high-dimensional clinical data to forecast postoperative complications","volume":"11","author":"Thottakkara","year":"2016","journal-title":"PLoS One"},{"key":"B35","doi-asserted-by":"publisher","first-page":"2374289519873088","DOI":"10.1177\/2374289519873088","article-title":"Artificial intelligence and machine learning in pathology: the present landscape of supervised methods","volume":"6","author":"Rashidi","year":"2019","journal-title":"Acad Pathol"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2025.1583399\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,12]],"date-time":"2025-05-12T05:39:38Z","timestamp":1747028378000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2025.1583399\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,12]]},"references-count":35,"alternative-id":["10.3389\/fdgth.2025.1583399"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2025.1583399","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,12]]},"article-number":"1583399"}}