{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T19:03:39Z","timestamp":1782846219099,"version":"3.54.5"},"reference-count":47,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Form Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Accurately differentiating severe from nonsevere COVID-19 clinical types is critical for the health care system to optimize workflow. Current techniques lack the ability to accurately classify COVID-19 clinical types in patients, especially as SARS-CoV-2 continues to mutate.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n We explore the predictability and interpretability of multiple state-of-the-art machine learning (ML) techniques trained and tested under different biomedical data types and SARS-CoV-2 variants.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n Comprehensive patient-level data were collected from 362 patients (severe COVID-19: n=148; nonsevere COVID-19: n=214) infected with the original SARS-CoV-2 strain in 2020 and 1000 patients (severe COVID-19: n=500; nonsevere COVID-19: n=500) infected with the Omicron variant in 2022\u20102023. The data included 26 biochemical features from blood testing and 26 clinical features from patients\u2019 clinical characteristics and medical history. Different ML techniques, including penalized logistic regression, random forest,\n k<\/jats:italic>\n -nearest neighbors, and support vector machines, were applied to build predictive classification models based on each data modality separately and together for each variant. Fifty randomized train-test splits were conducted per scenario, and performance results were recorded.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n The fusion (hybrid) characteristic modality yielded the highest mean area under the curve (AUC) in this study, achieving 0.915, while the biochemical and clinical modalities had AUCs of 0.862 and 0.818, respectively. All ML models performed similarly under different testing scenarios and were consistent when cross-tested with data of patients infected with the original strain and those infected with the Omicron variant. Our models ranked elevated d-dimer (biochemical), elevated high sensitivity troponin I (biochemical), and age greater than 55 years (clinical) as the most positively predictive features of severe COVID-19.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n These results are compatible with the hypothesis that ML is a useful tool for predicting severe COVID-19 based on comprehensive individual patient\u2013level data. Further, ML models trained on the biochemical and clinical modalities together show patterns consistent with enhanced predictive performance. The improved performance observed with Omicron variant data agrees with the hypothesis that ML approaches may retain utility across variants in this study setting, although further validation is required before clinical application. Future work using larger datasets with more ethnic variation and investigating unbiased ML interpretation methods may be able to provide further validation.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/76542","type":"journal-article","created":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T17:21:16Z","timestamp":1777569676000},"page":"e76542-e76542","source":"Crossref","is-referenced-by-count":0,"title":["Evaluating Biomedical Feature Fusion on Machine Learning\u2019s Predictability and Interpretability of COVID-19 Severity Types: Model Development, Interpretation, and Validation"],"prefix":"10.2196","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-3787-8960","authenticated-orcid":false,"given":"Haleigh Noelle","family":"West-Page","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9665-126X","authenticated-orcid":false,"given":"Kevin","family":"McGoff","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-6464-9247","authenticated-orcid":false,"given":"Harrison","family":"Latimer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9480-4813","authenticated-orcid":false,"given":"Isaac","family":"Olufadewa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2316-111X","authenticated-orcid":false,"given":"Shi","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,4,30]]},"reference":[{"key":"R1","doi-asserted-by":"publisher","DOI":"10.1016\/j.envres.2022.112816","article-title":"Emergence of SARS-CoV-2 Omicron (B.1.1.529) variant, salient features, high global health concerns and strategies to counter it amid ongoing COVID-19 pandemic","volume":"209","author":"Khandia","journal-title":"Environ Res"},{"issue":"6","key":"R2","doi-asserted-by":"publisher","first-page":"667","DOI":"10.1097\/MCC.0000000000000993","article-title":"Impact of ICU strain on outcomes","volume":"28","author":"Duggal","journal-title":"Curr Opin Crit Care"},{"issue":"4","key":"R3","doi-asserted-by":"publisher","first-page":"211","DOI":"10.12788\/jhm.3539","article-title":"Analysis of hospital resource availability and COVID-19 mortality across the United States","volume":"16","author":"Janke","journal-title":"J Hosp Med"},{"key":"R4","unstructured":"COVID-19 epidemiological update \u2013 16 February 2024. World Health Organization. 2024. URL: https:\/\/www.who.int\/publications\/m\/item\/covid-19-epidemiological-update-16-february-2024 [Accessed 16-02-2024]"},{"issue":"4","key":"R5","doi-asserted-by":"publisher","DOI":"10.2196\/23948","article-title":"A multimodality machine learning approach to differentiate severe and nonsevere COVID-19: Model development and validation","volume":"23","author":"Chen","journal-title":"J Med Internet Res"},{"issue":"13","key":"R6","doi-asserted-by":"publisher","first-page":"1239","DOI":"10.1001\/jama.2020.2648","article-title":"Characteristics of and important lessons from the coronavirus disease 2019 (COVID-19) outbreak in China","volume":"323","author":"Wu","journal-title":"JAMA"},{"issue":"17","key":"R7","doi-asserted-by":"publisher","DOI":"10.3390\/healthcare13172154","article-title":"Explainable AI in clinical decision support systems: A meta-analysis of methods, applications, and usability challenges","volume":"13","author":"Abbas","journal-title":"Healthcare (Basel)"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.1016\/j.imu.2024.101587","article-title":"A survey of explainable artificial intelligence in healthcare: Concepts, applications, and challenges","volume":"51","author":"Mienye","journal-title":"Informatics in Medicine Unlocked"},{"issue":"22","key":"R9","doi-asserted-by":"publisher","first-page":"15693","DOI":"10.1007\/s00521-021-06189-y","article-title":"SMOTE-NC and gradient boosting imputation based random forest classifier for predicting severity level of covid-19 patients with blood samples","volume":"33","author":"G\u00f6k","journal-title":"Neural Comput Appl"},{"issue":"6","key":"R10","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0253329","article-title":"The selection of indicators from initial blood routine test results to improve the accuracy of early prediction of COVID-19 severity","volume":"16","author":"Luo","journal-title":"PLoS ONE"},{"issue":"1","key":"R11","doi-asserted-by":"publisher","DOI":"10.1186\/s40249-022-00946-4","article-title":"Comparing different machine learning techniques for predicting COVID-19 severity","volume":"11","author":"Xiong","journal-title":"Infect Dis Poverty"},{"key":"R12","unstructured":"Novel Coronavirus Pneumonia Diagnosis and Treatment Plan (Provisional 7th Edition). 7th ed. National Health Commission of China; 2020. URL: http:\/\/www.nhc.gov.cn\/yzygj\/s7652m\/202003\/a31191442e29474b98bfed5579d5af95.shtml [Accessed 20-04-2026]"},{"issue":"7","key":"R13","doi-asserted-by":"publisher","first-page":"e45","DOI":"10.1164\/rccm.201908-1581ST","article-title":"Diagnosis and treatment of adults with community-acquired pneumonia. An official clinical practice guideline of the American Thoracic Society and Infectious Diseases Society of America","volume":"200","author":"Metlay","journal-title":"Am J Respir Crit Care Med"},{"issue":"6","key":"R14","doi-asserted-by":"publisher","first-page":"6102","DOI":"10.3934\/mbe.2022285","article-title":"A new hybrid ensemble machine-learning model for severity risk assessment and post-COVID prediction system","volume":"19","author":"Shakhovska","journal-title":"Math Biosci Eng"},{"issue":"2","key":"R15","doi-asserted-by":"publisher","first-page":"421","DOI":"10.1515\/cclm-2020-1294","article-title":"Development, evaluation, and validation of machine learning models for COVID-19 detection based on routine blood tests","volume":"59","author":"Cabitza","journal-title":"Clin Chem Lab Med"},{"issue":"15","key":"R16","doi-asserted-by":"publisher","DOI":"10.3390\/s19153400","article-title":"Logistic regression for machine learning in process tomography","volume":"19","author":"Rymarczyk","journal-title":"Sensors (Basel)"},{"issue":"6","key":"R17","doi-asserted-by":"publisher","first-page":"6413","DOI":"10.1007\/s10489-021-02743-2","article-title":"Machine learning techniques to predict different levels of hospital care of CoVid-19","volume":"52","author":"Hern\u00e1ndez-Pereira","journal-title":"Appl Intell (Dordr)"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.3389\/fdgth.2021.681608","article-title":"Using machine learning to predict mortality for COVID-19 patients on day 0 in the ICU","volume":"3","author":"Jamshidi","journal-title":"Front Digit Health"},{"issue":"3","key":"R19","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0247773","article-title":"Clinical decision support tool for diagnosis of COVID-19 in hospitals","volume":"16","author":"Saegerman","journal-title":"PLOS ONE"},{"issue":"3","key":"R20","doi-asserted-by":"publisher","first-page":"148","DOI":"10.4103\/picr.PICR_87_17","article-title":"Common pitfalls in statistical analysis: Logistic regression","volume":"8","author":"Ranganathan","journal-title":"Perspect Clin Res"},{"issue":"1","key":"R21","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1177\/1536867X20909688","article-title":"The random forest algorithm for statistical learning","volume":"20","author":"Schonlau","journal-title":"Stata J"},{"issue":"11","key":"R22","doi-asserted-by":"publisher","first-page":"218","DOI":"10.21037\/atm.2016.03.37","article-title":"Introduction to machine learning: k-nearest neighbors","volume":"4","author":"Zhang","journal-title":"Ann Transl Med"},{"key":"R23","doi-asserted-by":"publisher","DOI":"10.1016\/j.dajour.2022.100071","article-title":"A comparative analysis of K-Nearest Neighbor, Genetic, Support Vector Machine, Decision Tree, and Long Short Term Memory algorithms in machine learning","volume":"3","author":"Bansal","journal-title":"Decision Analytics Journal"},{"issue":"1","key":"R24","doi-asserted-by":"publisher","DOI":"10.1186\/s13104-020-05180-5","article-title":"A decision support system based on support vector machine for diagnosis of periodontal disease","volume":"13","author":"Farhadian","journal-title":"BMC Res Notes"},{"key":"R25","doi-asserted-by":"publisher","DOI":"10.12688\/f1000research.20498.2","article-title":"Evidence-based clinical decision support systems for the prediction and detection of three disease states in critical care: A systematic literature review","volume":"8","author":"Medic","journal-title":"F1000Res"},{"issue":"85","key":"R26","first-page":"2825","author":"Pedregosa","journal-title":"J Mach Learn Res"},{"key":"R27","doi-asserted-by":"crossref","unstructured":"Aliferis C Simon G . Overfitting, Underfitting and General Model Overconfidence and Under-Performance Pitfalls and Best Practices in Machine Learning and AI. In: Simon GJ Aliferis C , editors. Artificial Intelligence and Machine Learning in Health Care and Medical Sciences: Best Practices and Pitfalls. 2024:477-524. [doi: 10.1007\/978-3-031-39355-6_10] [Medline: 39836821]","DOI":"10.1007\/978-3-031-39355-6_10"},{"issue":"3","key":"R28","doi-asserted-by":"publisher","first-page":"299","DOI":"10.1109\/TKDE.2005.50","article-title":"Using AUC and accuracy in evaluating learning algorithms","volume":"17","author":"Ling","journal-title":"IEEE Trans Knowl Data Eng"},{"issue":"2","key":"R29","doi-asserted-by":"publisher","DOI":"10.1007\/s42452-021-04148-9","article-title":"Comparison of feature importance measures as explanations for classification models","volume":"3","author":"Saarela","journal-title":"SN Appl Sci"},{"key":"R30","unstructured":"Feature importances with a forest of trees. scikit-learn. 2024. URL: https:\/\/scikit-learn.org\/1.5\/auto_examples\/ensemble\/plot_forest_importances.html [Accessed 20-04-2026]"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.1711","article-title":"Feature selection based on neighborhood rough sets and Gini index","volume":"9","author":"Zhang","journal-title":"PeerJ Comput Sci"},{"issue":"12","key":"R32","doi-asserted-by":"publisher","DOI":"10.2196\/26323","article-title":"Value of the electronic medical record for hospital care: Update from the literature","volume":"23","author":"Uslu","journal-title":"J Med Internet Res"},{"issue":"7","key":"R33","doi-asserted-by":"publisher","first-page":"802","DOI":"10.1001\/jamacardio.2020.0950","article-title":"Association of cardiac injury with mortality in hospitalized patients with COVID-19 in Wuhan, China","volume":"5","author":"Shi","journal-title":"JAMA Cardiol"},{"issue":"195","key":"R34","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1016\/j.thromres.2020.07.047","article-title":"D-dimer level is associated with the severity of COVID-19","volume":"195","author":"Yu","journal-title":"Thromb Res"},{"key":"R35","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2020.100448","article-title":"Characteristics, risk factors and outcomes among the first consecutive 1096 patients diagnosed with COVID-19 in Kuwait","volume":"24","author":"Almazeedi","journal-title":"EClinicalMedicine"},{"issue":"6","key":"R36","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2020.12270","article-title":"Clinical characteristics and morbidity associated with coronavirus disease 2019 in a series of patients in Metropolitan Detroit","volume":"3","author":"Suleyman","journal-title":"JAMA Netw Open"},{"issue":"10","key":"R37","doi-asserted-by":"publisher","first-page":"e516","DOI":"10.1016\/S2589-7500(20)30217-X","article-title":"Clinical features of COVID-19 mortality: development and validation of a clinical prediction model","volume":"2","author":"Yadaw","journal-title":"Lancet Digit Health"},{"issue":"9","key":"R38","doi-asserted-by":"publisher","first-page":"1830","DOI":"10.1093\/aje\/kwab010","article-title":"Addressing measurement error in random forests using quantitative bias analysis","volume":"190","author":"Jiang","journal-title":"Am J Epidemiol"},{"issue":"1","key":"R39","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1111\/cts.12884","article-title":"Precision medicine, AI, and the future of personalized health care","volume":"14","author":"Johnson","journal-title":"Clin Transl Sci"},{"issue":"6","key":"R40","doi-asserted-by":"publisher","first-page":"383","DOI":"10.1002\/hcs2.122","article-title":"Innovative public strategies in response to COVID-19: A review of practices from China","volume":"3","author":"Wu","journal-title":"Health Care Sci"},{"key":"R41","unstructured":"China\u2019s actions to combat the new coronavirus pneumonia outbreak. Information Office of the State Council of the People\u2019s Republic of China; URL: http:\/\/www.gov.cn\/zhengce\/2020-06\/07\/content_5517737.htm [Accessed 12-05-2023]"},{"issue":"3","key":"R42","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1002\/hcs2.45","article-title":"Evolution and major changes of the diagnosis and treatment protocol for COVID-19 patients in China 2020-2023","volume":"2","author":"Wu","journal-title":"Health Care Sci"},{"issue":"21","key":"R43","doi-asserted-by":"publisher","first-page":"3711","DOI":"10.1093\/bioinformatics\/bty373","article-title":"The revival of the Gini importance?","volume":"34","author":"Nembrini","journal-title":"Bioinformatics"},{"key":"R44","doi-asserted-by":"publisher","DOI":"10.1155\/2015\/471371","article-title":"Unbiased feature selection in learning random forests for high-dimensional data","volume":"2015","author":"Nguyen","journal-title":"ScientificWorldJournal"},{"issue":"5","key":"R45","doi-asserted-by":"publisher","DOI":"10.3390\/healthcare13050507","article-title":"HeartEnsembleNet: An innovative hybrid ensemble learning approach for cardiovascular risk prediction","volume":"13","author":"Zaidi","journal-title":"Healthcare (Basel)"},{"issue":"13","key":"R46","doi-asserted-by":"publisher","DOI":"10.3390\/math13132194","article-title":"A novel approach utilizing bagging, histogram gradient boosting, and advanced feature selection for predicting the onset of cardiovascular diseases","volume":"13","author":"Fitriyani","journal-title":"Mathematics"},{"key":"R47","unstructured":"Hnwestpage\/fusion-ML-COVID-19. GitHub. URL: https:\/\/github.com\/hnwestpage\/Fusion-ML-COVID-19 [Accessed 20-04-2026]"}],"container-title":["JMIR Formative Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e76542","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:08:08Z","timestamp":1782842888000},"score":1,"resource":{"primary":{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e76542"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,30]]},"references-count":47,"URL":"https:\/\/doi.org\/10.2196\/76542","relation":{},"ISSN":["2561-326X"],"issn-type":[{"value":"2561-326X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,30]]},"article-number":"v10i6e76542"}}