{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T18:09:23Z","timestamp":1773425363509,"version":"3.50.1"},"reference-count":32,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T00:00:00Z","timestamp":1773360000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"\n Background<\/jats:title>\n The COVID-19 pandemic has highlighted the critical need for robust, interpretable predictive models to guide clinical decision-making for hospitalized patients, particularly in resource-limited settings. While machine learning approaches have achieved high predictive performance, many lack the transparency required for clinical adoption. Bayesian networks provide a rigorous mathematical framework for modeling medical uncertainty and complex causal relationships while maintaining clinical interpretability.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n To develop and validate an interpretable Bayesian network model for predicting clinical outcomes in hospitalized COVID-19 patients, including severity, complications, and mortality, and to compare its performance with existing predictive approaches. The model specifically addresses the needs of resource-limited settings where both interpretability and performance are critical.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n This retrospective cohort study analyzed 124 hospitalized COVID-19 patients at Tanambao I University Hospital, Madagascar (March 2020\u2013March 2022). Given limited genomic sequencing capacity during the study period, specific SARS-CoV-2 variants infecting individual patients could not be confirmed. After descriptive and statistical implicative analysis, we constructed a Bayesian network integrating predictive variables organized in three levels: input variables (demographic, vital signs, biological, radiological), intermediate variables (clinical scores), and target variables (severity, complications, evolution, death). Parameter learning used maximum likelihood estimation with Laplace smoothing. Model performance was evaluated using stratified 10-fold cross-validation and compared against logistic regression, random forest, and support vector machine approaches.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n The Bayesian network model exhibited strong diagnostic performance, with an AUC of 0.95 (95% CI: 0.91\u20130.99) for death prediction, 0.94 (95% CI: 0.90\u20130.98) for severe outcomes, and 0.93 (95% CI: 0.89\u20130.97) for unfavorable progression. Stratified 10-fold cross-validation yielded a mean accuracy of\n \n \n \n \n \n 0.85<\/mml:mn>\n \u00b1<\/mml:mi>\n 0.03<\/mml:mn>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:alternatives>\n <\/jats:inline-formula>\n . The model outperformed logistic regression (AUC 0.89), random forest (AUC 0.91), and SVM (AUC 0.87) while maintaining superior interpretability. Sensitivity analysis identified qSOFA score (mutual information: 0.342), SpO\n 2<\/jats:sub>\n levels (0.298), and respiratory distress (0.276) as the most influential variables. Robustness testing showed prediction stability under parameter perturbations of\n \n \n \n \n \n \u00b1<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:alternatives>\n <\/jats:inline-formula>\n 15%, with variation remaining below 8%.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Bayesian networks constitute a promising tool for COVID-19 outcome prediction in resource-limited settings, combining competitive predictive performance with essential clinical interpretability. The probabilistic approach enables rigorous uncertainty quantification critical for medical decision-making. While the model achieved excellent performance on our cohort, external validation across different populations and SARS-CoV-2 variants is needed to establish broader generalizability.<\/jats:p>\n <\/jats:sec>","DOI":"10.1371\/journal.pone.0343096","type":"journal-article","created":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T17:37:11Z","timestamp":1773423431000},"page":"e0343096","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Bayesian networks for predicting clinical outcomes in COVID-19 patients: A retrospective study in a resource-limited setting"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-5978-0709","authenticated-orcid":true,"given":"Tombolaza Canut","family":"Filamant","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Angelo Fulgence","family":"Raherinirina","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andr\u00e9","family":"Totohasina","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2026,3,13]]},"reference":[{"issue":"10229","key":"pone.0343096.ref001","doi-asserted-by":"crossref","first-page":"1054","DOI":"10.1016\/S0140-6736(20)30566-3","article-title":"Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study","volume":"395","author":"F Zhou","year":"2020","journal-title":"Lancet"},{"issue":"18","key":"pone.0343096.ref002","doi-asserted-by":"crossref","first-page":"1708","DOI":"10.1056\/NEJMoa2002032","article-title":"Clinical Characteristics of Coronavirus Disease 2019 in China","volume":"382","author":"W-J Guan","year":"2020","journal-title":"N Engl J Med"},{"key":"pone.0343096.ref003","article-title":"Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal","volume":"369","author":"L Wynants","year":"2020","journal-title":"BMJ"},{"issue":"5","key":"pone.0343096.ref004","doi-asserted-by":"crossref","first-page":"283","DOI":"10.1038\/s42256-020-0180-7","article-title":"An interpretable mortality prediction model for COVID-19 patients","volume":"2","author":"L Yan","year":"2020","journal-title":"Nat Mach Intell"},{"issue":"1","key":"pone.0343096.ref005","doi-asserted-by":"crossref","first-page":"328","DOI":"10.1186\/s13054-021-03749-5","article-title":"Machine-learning-based COVID-19 mortality prediction model and identification of patients at low and high risk of dying","volume":"25","author":"MM Banoei","year":"2021","journal-title":"Crit Care"},{"issue":"14","key":"pone.0343096.ref006","doi-asserted-by":"crossref","first-page":"1347","DOI":"10.1056\/NEJMra1814259","article-title":"Machine learning in medicine","volume":"380","author":"A Rajkomar","year":"2019","journal-title":"N Engl J Med"},{"key":"pone.0343096.ref007","volume-title":"Probabilistic reasoning in intelligent systems: networks of plausible inference","author":"J Pearl","year":"1988"},{"issue":"3","key":"pone.0343096.ref008","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1016\/j.artmed.2003.11.001","article-title":"Bayesian networks in biomedicine and health-care","volume":"30","author":"PJF Lucas","year":"2004","journal-title":"Artif Intell Med"},{"key":"pone.0343096.ref009","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4614-6446-4","volume-title":"Bayesian networks in R: with applications in systems biology","author":"R Nagarajan","year":"2013"},{"key":"pone.0343096.ref010","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1007\/BF03190296","article-title":"A Bayesian network for diagnosis of primary bone tumors","volume":"14","author":"CE Jr Kahn","year":"2001","journal-title":"J Digit Imaging"},{"issue":"3","key":"pone.0343096.ref011","doi-asserted-by":"crossref","first-page":"944","DOI":"10.1109\/JBHI.2015.2420534","article-title":"DBN-Extended: A Dynamic Bayesian Network Model Extended With Temporal Abstractions for Coronary Heart Disease Prognosis","volume":"20","author":"K Orphanou","year":"2016","journal-title":"IEEE J Biomed Health Inform"},{"key":"pone.0343096.ref012","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-540-78983-3","volume-title":"Statistical implication analysis: theory and applications","author":"R Gras","year":"2008"},{"issue":"15","key":"pone.0343096.ref013","doi-asserted-by":"crossref","first-page":"1499","DOI":"10.1001\/jama.2020.3633","article-title":"Care for Critically Ill Patients With COVID-19","volume":"323","author":"S Murthy","year":"2020","journal-title":"JAMA"},{"issue":"10","key":"pone.0343096.ref014","doi-asserted-by":"crossref","DOI":"10.1016\/S2214-109X(21)00361-2","article-title":"Seroprevalence of ancestral and Beta SARS-CoV-2 antibodies in Malagasy blood donors","volume":"9","author":"SL Razafimahatratra","year":"2021","journal-title":"Lancet Glob Health"},{"issue":"6566","key":"pone.0343096.ref015","doi-asserted-by":"crossref","first-page":"423","DOI":"10.1126\/science.abj4336","article-title":"A year of genomic surveillance reveals how the SARS-CoV-2 pandemic unfolded in Africa","volume":"374","author":"E Wilkinson","year":"2021","journal-title":"Science"},{"issue":"10332","key":"pone.0343096.ref016","doi-asserted-by":"crossref","first-page":"1303","DOI":"10.1016\/S0140-6736(22)00462-7","article-title":"Comparative analysis of the risks of hospitalisation and death associated with SARS-CoV-2 omicron (B.1.1.529) and delta (B.1.617.2) variants in England: a cohort study","volume":"399","author":"T Nyberg","year":"2022","journal-title":"Lancet"},{"issue":"10323","key":"pone.0343096.ref017","doi-asserted-by":"crossref","first-page":"437","DOI":"10.1016\/S0140-6736(22)00017-4","article-title":"Early assessment of the clinical severity of the SARS-CoV-2 omicron variant in South Africa: a data linkage study","volume":"399","author":"N Wolter","year":"2022","journal-title":"Lancet"},{"issue":"8","key":"pone.0343096.ref018","doi-asserted-by":"crossref","first-page":"801","DOI":"10.1001\/jama.2016.0287","article-title":"The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3)","volume":"315","author":"M Singer","year":"2016","journal-title":"JAMA"},{"key":"pone.0343096.ref019","volume-title":"Causation, prediction, and search","author":"P Spirtes","year":"2000"},{"issue":"3","key":"pone.0343096.ref020","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v035.i03","article-title":"Learning Bayesian networks with the bnlearn R package","volume":"35","author":"M Scutari","year":"2010","journal-title":"J Stat Softw"},{"issue":"4","key":"pone.0343096.ref021","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1214\/ss\/1177011136","article-title":"Inference from iterative simulation using multiple sequences","volume":"7","author":"A Gelman","year":"1992","journal-title":"Stat Sci"},{"issue":"3","key":"pone.0343096.ref022","doi-asserted-by":"crossref","first-page":"837","DOI":"10.2307\/2531595","article-title":"Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach","volume":"44","author":"ER DeLong","year":"1988","journal-title":"Biometrics"},{"key":"pone.0343096.ref023","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"pone.0343096.ref024","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1186\/1471-2105-12-77","article-title":"pROC: an open-source package for R and S+ to analyze and compare ROC curves","volume":"12","author":"X Robin","year":"2011","journal-title":"BMC Bioinformatics"},{"issue":"5","key":"pone.0343096.ref025","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v028.i05","article-title":"Building predictive models in R using the caret package","volume":"28","author":"M Kuhn","year":"2008","journal-title":"J Stat Softw"},{"key":"pone.0343096.ref026","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-24277-4","volume-title":"ggplot2: elegant graphics for data analysis","author":"H Wickham","year":"2016"},{"key":"pone.0343096.ref027","article-title":"Risk stratification of patients admitted to hospital with covid-19 using the ISARIC WHO Clinical Characterisation Protocol: development and validation of the 4C Mortality Score","volume":"370","author":"SR Knight","year":"2020","journal-title":"BMJ"},{"issue":"14","key":"pone.0343096.ref028","doi-asserted-by":"crossref","first-page":"1469","DOI":"10.1001\/jama.2021.1545","article-title":"Discriminant accuracy of the SOFA score for determining the probable mortality of patients with COVID-19 pneumonia requiring mechanical ventilation","volume":"325","author":"RA Raschke","year":"2021","journal-title":"JAMA"},{"key":"pone.0343096.ref029","volume-title":"Dynamic Bayesian networks: representation, inference and learning","author":"KP Murphy","year":"2002"},{"issue":"10178","key":"pone.0343096.ref030","doi-asserted-by":"crossref","first-page":"1331","DOI":"10.1016\/S0140-6736(19)30024-8","article-title":"Building a tuberculosis-free world: The Lancet Commission on tuberculosis","volume":"393","author":"MJA Reid","year":"2019","journal-title":"Lancet"},{"key":"pone.0343096.ref031","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511803161","volume-title":"Causality: models, reasoning, and inference","author":"J Pearl","year":"2009"},{"issue":"9840","key":"pone.0343096.ref032","doi-asserted-by":"crossref","first-page":"507","DOI":"10.1016\/S0140-6736(12)61127-1","article-title":"Technologies for global health","volume":"380","author":"P Howitt","year":"2012","journal-title":"Lancet"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0343096","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,13]],"date-time":"2026-03-13T17:37:18Z","timestamp":1773423438000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0343096"}},"subtitle":[],"editor":[{"given":"Ramya","family":"Iyadurai","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,3,13]]},"references-count":32,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3,13]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0343096","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,13]]}}}