{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T05:09:52Z","timestamp":1779253792366,"version":"3.51.4"},"reference-count":37,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T00:00:00Z","timestamp":1779235200000},"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. Med."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Heart failure (HF) is a major public health issue globally, requiring early risk stratification to improve patient outcomes. Despite numerous machine learning (ML) applications in HF prediction, critical gaps persist: lack of systematic algorithm benchmarking under standardized conditions, reliance on single explainable AI (XAI) methods, and insufficient attention to class imbalance. We provide the first comprehensive benchmark of 10 ML algorithms for HF risk stratification and establish a validated dual-XAI framework for clinical deployment.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We analyzed a publicly available dataset of 2,169 patients with 15 clinical features. Ten ML algorithms (five traditional: Logistic Regression, Decision Tree, Support Vector Machine, K-Nearest Neighbors, Naive Bayes; five ensemble: Random Forest, Gradient Boosting, XGBoost, LightGBM, CatBoost) were compared under rigorously standardized conditions, including an 80:20 stratified train-test split, identical preprocessing, and a fixed random seed. Performance was evaluated using six complementary metrics: Receiver Operating Characteristic-Area Under Curve (ROC-AUC), accuracy, precision, recall, F1-score, and Precision-Recall AUC (PR-AUC). SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) were applied to interpret the best model, with quantitative cross-method concordance analysis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Logistic Regression achieved optimal performance (ROC-AUC = 0.9451, accuracy = 88.25%, F1-score = 0.8294, PR-AUC = 0.9113), outperforming complex ensemble methods. Left Ventricular Ejection Fraction (LVEF) was the dominant predictor (SHAP normalized importance = 1.0; LIME weight = 0.2917), followed by diabetes, age, hypertension, and serum creatinine, with 100% SHAP-LIME concordance in top-3 rankings. The false-negative rate on the test set was 18.42% (28\/152), supporting the model's clinical utility for high-risk patient identification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Key contributions include: (1) the first systematic comparison of 10 algorithms under identical experimental conditions; (2) a novel dual-XAI framework (SHAP + LIME) with quantified cross-validation, addressing single-method biases; (3) demonstration that simple, interpretable models outperform complex ensembles on moderate-sized datasets (n = 2,169), offering practical guidance for resource-limited settings; and (4) quantitative validation of LVEF dominance across interpretation methods. The validated framework shows potential for clinical risk stratification systems, though prospective validation against independent clinical outcomes is required.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1801693","type":"journal-article","created":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T04:44:28Z","timestamp":1779252268000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Heart failure risk prediction based on machine learning and interpretability analysis"],"prefix":"10.3389","volume":"13","author":[{"given":"Hangqian","family":"Li","sequence":"first","affiliation":[{"name":"Department of Informatics, Computer Science, King's College London","place":["London, United Kingdom"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,20]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"81542","DOI":"10.1109\/ACCESS.2019.2923707","article-title":"Effective heart disease prediction using hybrid machine learning techniques","volume":"7","author":"Mohan","year":"2019","journal-title":"IEEE Access"},{"key":"B2","doi-asserted-by":"publisher","first-page":"163","DOI":"10.1016\/j.eswa.2016.10.020","article-title":"An integrated decision support system based on ANN and Fuzzy_AHP for heart failure risk prediction","volume":"68","author":"Samuel","year":"2017","journal-title":"Expert Syst Appl"},{"key":"B3","doi-asserted-by":"publisher","first-page":"100402","DOI":"10.1016\/j.imu.2020.100402","article-title":"An improved ensemble learning approach for the prediction of heart disease risk","volume":"20","author":"Mienye","year":"2020","journal-title":"Inf Med Unlocked"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1161","DOI":"10.3390\/s23031161","article-title":"Efficient data-driven machine learning models for cardiovascular diseases risk prediction","volume":"23","author":"Dritsas","year":"2023","journal-title":"Sensors"},{"key":"B5","doi-asserted-by":"publisher","first-page":"100203","DOI":"10.1016\/j.imu.2019.100203","article-title":"Improving the accuracy of prediction of heart disease risk based on ensemble classification techniques","volume":"16","author":"Latha","year":"2019","journal-title":"Inf Med Unlocked"},{"key":"B6","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1016\/j.future.2019.10.043","article-title":"HealthFog: an ensemble deep learning based smart healthcare system for automatic diagnosis of heart diseases in integrated IoT and fog computing environments","volume":"104","author":"Tuli","year":"2020","journal-title":"Future Gener Comput Syst"},{"key":"B7","doi-asserted-by":"publisher","first-page":"i2416","DOI":"10.1136\/bmj.i2416","article-title":"Prediction models for cardiovascular disease risk in the general population: systematic review","volume":"353","author":"Damen","year":"2016","journal-title":"BMJ"},{"key":"B8","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1016\/j.inffus.2019.12.012","article-title":"Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI","volume":"58","author":"Barredo Arrieta","year":"2020","journal-title":"Inf Fusion"},{"key":"B9","doi-asserted-by":"publisher","first-page":"e0231166","DOI":"10.1371\/journal.pone.0231166","article-title":"Opening the black box of artificial intelligence for clinical decision support: a study predicting stroke outcome","volume":"15","author":"Zihni","year":"2020","journal-title":"PLoS ONE"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1384","DOI":"10.1109\/TPWRS.2023.3248941","article-title":"Using SHAP values and machine learning to understand trends in the transient stability limit","volume":"39","author":"Hamilton","year":"2024","journal-title":"IEEE Trans Power Syst"},{"key":"B11","doi-asserted-by":"publisher","first-page":"2681","DOI":"10.3390\/diagnostics13162681","article-title":"Interpretable machine learning for personalized medical recommendations: a LIME-based approach","volume":"13","author":"Wu","year":"2023","journal-title":"Diagnostics"},{"key":"B12","doi-asserted-by":"publisher","first-page":"78994","DOI":"10.1109\/ACCESS.2023.3294569","article-title":"A review of trustworthy and explainable artificial intelligence (XAI)","volume":"11","author":"Chamola","year":"2023","journal-title":"IEEE Access"},{"key":"B13","doi-asserted-by":"publisher","first-page":"2184","DOI":"10.1016\/j.eswa.2014.10.027","article-title":"Classification of healthcare data using genetic fuzzy logic system and wavelets","volume":"42","author":"Nguyen","year":"2015","journal-title":"Expert Syst Appl"},{"key":"B14","doi-asserted-by":"publisher","first-page":"129","DOI":"10.32604\/jai.2024.050277","article-title":"Cardiovascular disease prediction using risk factors: a comparative performance analysis of machine learning models","volume":"6","author":"Hussain","year":"2024","journal-title":"J Artif Intell"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/eSmarTA66764.2025.11132291","article-title":"Heart disease classification using machine learning techniques: an impact analysis of GridSearchCV-based optimization","author":"Hussain","year":"2025","journal-title":"2025 5th international conference on emerging smart technologies and applications (eSmarTA)"},{"key":"B16","doi-asserted-by":"publisher","first-page":"208","DOI":"10.1016\/j.inffus.2020.06.008","article-title":"A smart healthcare monitoring system for heart disease prediction based on ensemble deep learning and feature fusion","volume":"63","author":"Ali","year":"2022","journal-title":"Inf Fusion"},{"key":"B17","doi-asserted-by":"publisher","first-page":"0075","DOI":"10.34133\/cbsystems.0075","article-title":"Learning representations from heart sound: a comparative study on shallow and deep models","volume":"5","author":"Qian","year":"2024","journal-title":"Cyborg Bionic Syst"},{"key":"B18","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1186\/s40537-024-01050-0","article-title":"Quantum computational infusion in extreme learning machines for early multi-cancer detection","volume":"12","author":"Bilal","year":"2025","journal-title":"J Big Data"},{"key":"B19","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1093\/eurjhf\/hft132","article-title":"Uric acid and risk of heart failure: a systematic review and meta-analysis","volume":"16","author":"Huang","year":"2014","journal-title":"Eur J Heart Fail"},{"key":"B20","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1002\/ehf2.14535","article-title":"Uric acid is a biomarker for heart failure, but not therapeutic target: result from a comprehensive meta-analysis","volume":"11","author":"Qin","year":"2024","journal-title":"ESC Heart Failure"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1146","DOI":"10.1016\/j.amjcard.2017.06.057","article-title":"Prevalence of hyperuricemia in patients with acute heart failure with either reduced or preserved ejection fraction","volume":"120","author":"Palazzuoli","year":"2017","journal-title":"Am J Cardiol"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1027","DOI":"10.1002\/ehf2.12765","article-title":"Prognostic impacts of serum uric acid levels in patients with chronic heart failure: insights from the CHART-2 study","volume":"8","author":"Fujihashi","year":"2021","journal-title":"ESC Heart Fail"},{"key":"B23","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1016\/j.ejim.2021.04.001","article-title":"Serum uric acid and outcomes in patients with chronic heart failure through the whole spectrum of ejection fraction phenotypes: analysis of the ESC-EORP heart failure long-term (HF LT) registry","volume":"89","author":"Ambrosio","year":"2021","journal-title":"Eur J Intern Med"},{"key":"B24","doi-asserted-by":"publisher","first-page":"362","DOI":"10.1053\/j.ajkd.2017.08.017","article-title":"Uric acid and the risks of kidney failure and death in individuals with CKD","volume":"71","author":"Srivastava","year":"2018","journal-title":"Am J Kidney Dis"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1674","DOI":"10.3390\/diagnostics11091674","article-title":"Hyperuricemia and progression of chronic kidney disease: a review from physiology and pathogenesis to the role of urate-lowering therapy","volume":"11","author":"Lee","year":"2021","journal-title":"Diagnostics"},{"key":"B26","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1177\/2047487320957793","article-title":"Role of comorbidities in heart failure prognosis part 2: chronic kidney disease, elevated serum uric acid","volume":"27","author":"Tedeschi","year":"2020","journal-title":"Eur J Prev Cardiol"},{"key":"B27","doi-asserted-by":"publisher","first-page":"e0169137","DOI":"10.1371\/journal.pone.0169137","article-title":"Increased serum sodium and serum osmolarity are independent risk factors for developing chronic kidney disease; 5 year cohort study","volume":"12","author":"Kuwabara","year":"2017","journal-title":"PLoS ONE"},{"key":"B28","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1161\/CIRCHEARTFAILURE.113.000529","article-title":"Effects of physical activity and sedentary time on the risk of heart failure","volume":"7","author":"Young","year":"2014","journal-title":"Circ Heart Fail"},{"key":"B29","doi-asserted-by":"publisher","first-page":"799","DOI":"10.1161\/CIRCRESAHA.118.312669","article-title":"Sedentary behavior, exercise, and cardiovascular health","volume":"124","author":"Lavie","year":"2019","journal-title":"Circ Res"},{"key":"B30","doi-asserted-by":"publisher","first-page":"291","DOI":"10.2147\/VHRM.S200278","article-title":"The association between sedentary behavior, physical activity and hyperuricemia","volume":"15","author":"Park","year":"2019","journal-title":"Vasc Health Risk Manag"},{"key":"B31","doi-asserted-by":"publisher","first-page":"701","DOI":"10.1161\/CIRCHEARTFAILURE.113.001010","article-title":"Dose-response relationship of total and leisure time physical activity to risk of heart failure: a prospective cohort study","volume":"7","author":"Andersen","year":"2014","journal-title":"Circ Heart Fail"},{"key":"B32","doi-asserted-by":"publisher","first-page":"s31","DOI":"10.1590\/1806-9282.66.s1.31","article-title":"Hyperkalemia in chronic kidney disease","volume":"66","author":"Watanabe","year":"2020","journal-title":"Rev Assoc M\u00e9d Bras"},{"key":"B33","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1186\/s12937-023-00888-z","article-title":"Potassium levels and the risk of all-cause and cardiovascular mortality among patients with cardiovascular diseases: a meta-analysis of cohort studies","volume":"23","author":"Fan","year":"2024","journal-title":"Nutr J"},{"key":"B34","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1177\/2048872615624241","article-title":"Serum potassium levels, cardiac arrhythmias, and mortality following non-ST-elevation myocardial infarction or unstable angina: insights from MERLIN-TIMI 36","volume":"6","author":"Patel","year":"2017","journal-title":"Eur Heart J Acute Cardiovasc Care"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1814","DOI":"10.1053\/euhj.1998.1188","article-title":"Uric acid in chronic heart failure: a marker of chronic inflammation","volume":"19","author":"Leyva","year":"1998","journal-title":"Eur Heart J"},{"key":"B36","doi-asserted-by":"publisher","first-page":"53","DOI":"10.3390\/life11010053","article-title":"Clinical implications of uric acid in heart failure: a comprehensive review","volume":"11","author":"Kumri\u0107","year":"2021","journal-title":"Life"},{"key":"B37","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1016\/j.ijcard.2015.08.089","article-title":"Uric acid and xanthine oxidase in heart failure\u2013emerging data and therapeutic implications","volume":"213","author":"Doehner","year":"2016","journal-title":"Int J Cardiol"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1801693\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T04:44:32Z","timestamp":1779252272000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1801693\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,20]]},"references-count":37,"alternative-id":["10.3389\/fmed.2026.1801693"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1801693","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,20]]},"article-number":"1801693"}}