{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T14:51:43Z","timestamp":1787755903106,"version":"build-2784847793"},"reference-count":32,"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 Patients\u2019 clinical status often evolves rapidly after an initial diagnosis, with each patient exhibiting a distinct disease trajectory. As a result, static risk scores fall short in supporting timely interventions\u2014an issue highlighted by COVID-19, where deaths have stemmed from heterogeneous pathways such as pneumonia, multiorgan failure, or exacerbation of preexisting conditions.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aims to propose a dynamic prognostic risk assessment framework based on longitudinal data collected during hospitalization, using COVID-19 as an example. Our aim was to develop and validate an interpretable framework that (1) screens prognosis at admission and (2) dynamically updates mortality risk throughout hospitalization, thereby providing clinicians with early, explainable warnings while minimizing additional cognitive load.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n In this retrospective study, we extracted electronic medical records of 382 COVID-19 cases treated at Tokyo Shinagawa Hospital between January 27 and September 30, 2020. At admission, gradient boosting decision trees (Light Gradient Boosting Machine) were used to predict the maximum clinical deterioration, including death, based on data available at initial diagnosis. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC). For in-hospital monitoring, random survival forests (RSF) were trained on a longitudinal dataset that combined static demographic characteristics with serially measured vital signs and laboratory results. The model dynamically assessed daily mortality risk by calculating a 7-day cumulative hazard function, with risk scores recalculated each day during hospitalization. RSF accuracy was evaluated in an independent one-third test set using the concordance index (C-index), an integrated Brier score (1-50 days), and mean time-dependent AUC. SurvSHAP(t), an extension of Shapley Additive Explanations, was applied to provide time-dependent explanations of each variable\u2019s contribution to the prediction.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n The prediction at initial diagnosis showed good agreement with the actual severity outcomes (AUC of 0.717 for predicting hospitalization\/severity \u22652; 0.878 for severity \u22653; 0.951 for severity \u22654; 0.952 for severity \u22655; and 0.970 for death\/severity=6), although some cases exhibited discrepancies between the predicted and actual prognoses. The dynamic mortality risk assessment during hospitalization using the RSF achieved a test-set C-index of 0.941, an integrated Brier score of 0.315, and a mean time-dependent AUC of 0.936. This dynamic assessment was able to distinguish between dead and surviving patients as early as 1\u20102 weeks before the outcome. Early in hospitalization, C-reactive protein was an important risk factor for mortality; during the middle period, peripheral oxygen saturation (SpO\n 2<\/jats:sub>\n ) gained importance; and immediately before death, platelets and \u03b2-D-glucan were the primary risk factors.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Integrating static admission triage with daily, explainable RSF predictions enables early identification of patients with COVID-19 at high risk of deterioration. By surfacing phase-specific, actionable predictors, the framework supports timely interventions and more efficient resource allocation. Prospective, multicenter studies are warranted to validate its generalizability and clinical impact.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/65585","type":"journal-article","created":{"date-parts":[[2025,5,28]],"date-time":"2025-05-28T10:40:07Z","timestamp":1748428807000},"page":"e65585-e65585","source":"Crossref","is-referenced-by-count":3,"title":["Explainable Machine Learning Framework for Dynamic Monitoring of Disease Prognostic Risk: Retrospective Cohort Study"],"prefix":"10.2196","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5418-5387","authenticated-orcid":false,"given":"Tetsuo","family":"Ishikawa","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0288-5637","authenticated-orcid":false,"given":"Masahiro","family":"Shinoda","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8016-2541","authenticated-orcid":false,"given":"Megumi","family":"Oya","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2292-9112","authenticated-orcid":false,"given":"Koichi","family":"Ashizaki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0922-8656","authenticated-orcid":false,"given":"Shinichiro","family":"Ota","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-9400-822X","authenticated-orcid":false,"given":"Kenichi","family":"Kamachi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4858-2607","authenticated-orcid":false,"given":"Kazuhiro","family":"Sakurada","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9955-4342","authenticated-orcid":false,"given":"Eiryo","family":"Kawakami","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4012-2518","authenticated-orcid":false,"given":"Masaharu","family":"Shinkai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,8,7]]},"reference":[{"issue":"3","key":"R1","doi-asserted-by":"publisher","first-page":"161","DOI":"10.7326\/0003-4819-138-3-200302040-00007","article-title":"The incidence and severity of adverse events affecting patients after discharge from the hospital","volume":"138","author":"Forster","journal-title":"Ann Intern Med"},{"issue":"10","key":"R2","doi-asserted-by":"publisher","first-page":"1992","DOI":"10.1111\/j.1365-2648.2009.05109.x","article-title":"Nurses\u2019 role in detecting deterioration in ward patients: systematic literature review","volume":"65","author":"Odell","journal-title":"J Adv Nurs"},{"key":"R3","doi-asserted-by":"publisher","DOI":"10.1111\/ijn.12329","article-title":"Vital signs monitoring to detect patient deterioration: an integrative literature review","volume":"21 Suppl 2","author":"Mok","journal-title":"Int J Nurs Pract"},{"issue":"1","key":"R4","doi-asserted-by":"publisher","DOI":"10.1186\/s13613-020-00689-1","article-title":"Management of critically ill patients with COVID-19 in ICU: statement from front-line intensive care experts in Wuhan, China","volume":"10","author":"Shang","journal-title":"Ann Intensive Care"},{"issue":"9749","key":"R5","doi-asserted-by":"publisher","first-page":"1339","DOI":"10.1016\/S0140-6736(10)60446-1","article-title":"Critical care and the global burden of critical illness in adults","volume":"376","author":"Adhikari","journal-title":"Lancet"},{"issue":"4","key":"R6","doi-asserted-by":"publisher","first-page":"872","DOI":"10.1093\/jamia\/ocv024","article-title":"Risk prediction for chronic kidney disease progression using heterogeneous electronic health record data and time series analysis","volume":"22","author":"Perotte","journal-title":"J Am Med Inform Assoc"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.1016\/j.lfs.2020.117839","article-title":"Coronaviruses pathogenesis, comorbidities and multi-organ damage - a review","volume":"255","author":"Renu","journal-title":"Life Sci"},{"issue":"2","key":"R8","doi-asserted-by":"publisher","first-page":"e78","DOI":"10.1016\/S2589-7500(19)30024-X","article-title":"Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records","volume":"1","author":"Nielsen","journal-title":"Lancet Digit Health"},{"issue":"2","key":"R9","doi-asserted-by":"publisher","first-page":"444","DOI":"10.1109\/JPROC.2015.2501978","article-title":"Machine learning and decision support in critical care","volume":"104","author":"Johnson","journal-title":"Proc IEEE Inst Electr Electron Eng"},{"issue":"7767","key":"R10","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1038\/s41586-019-1390-1","article-title":"A clinically applicable approach to continuous prediction of future acute kidney injury","volume":"572","author":"Toma\u0161ev","journal-title":"Nature New Biol"},{"issue":"4","key":"R11","doi-asserted-by":"publisher","first-page":"e179","DOI":"10.1016\/S2589-7500(20)30018-2","article-title":"Dynamic and explainable machine learning prediction of mortality in patients in the intensive care unit: a retrospective study of high-frequency data in electronic patient records","volume":"2","author":"Thorsen-Meyer","journal-title":"Lancet Digit Health"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2021.102158","article-title":"The three ghosts of medical AI: can the black-box present deliver?","volume":"124","author":"Quinn","journal-title":"Artif Intell Med"},{"key":"R13","unstructured":"Lundberg SM Lee SI . A unified approach to interpreting model predictions. Presented at: NIPS\u201917: Proceedings of the 31st International Conference on Neural Information Processing Systems; Dec 4-9, 2017:4768-4777; Long Beach, CA."},{"issue":"24","key":"R14","doi-asserted-by":"publisher","first-page":"2327","DOI":"10.1056\/NEJMoa2007016","article-title":"Compassionate use of remdesivir for patients with severe COVID-19","volume":"382","author":"Grein","journal-title":"N Engl J Med"},{"key":"R15","unstructured":"Ke G Meng Q Finley T et al. LightGBM: a highly efficient gradient boosting decision tree. In: Guyon I , editor. Advances in Neural Information Processing Systems. Vol 30. Curran Associates, Inc; 2017."},{"issue":"5","key":"R16","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy function approximation: A gradient boosting machine","volume":"29","author":"Friedman","journal-title":"Ann Statist"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1093\/bioinformatics\/btr597","article-title":"MissForest--non-parametric missing value imputation for mixed-type data","volume":"28","author":"Stekhoven","journal-title":"Bioinformatics"},{"key":"R18","doi-asserted-by":"crossref","unstructured":"Akiba T Sano S Yanase T et al. Optuna: a next-generation hyperparameter optimization framework. Presented at: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining; Aug 4-8, 2019:2623-2631; Anchorage, AK. [doi: 10.1145\/3292500.333070]","DOI":"10.1145\/3292500.3330701"},{"issue":"3","key":"R19","doi-asserted-by":"publisher","first-page":"841","DOI":"10.1214\/08-AOAS169","article-title":"Random survival forests","volume":"2","author":"Ishwaran","journal-title":"Ann Appl Stat"},{"key":"R20","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2022.110234","article-title":"SurvSHAP(t): time-dependent explanations of machine learning survival models","volume":"262","author":"Krzyzi\u0144ski","journal-title":"Knowl Based Syst"},{"issue":"5","key":"R21","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1038\/s42256-020-0180-7","article-title":"An interpretable mortality prediction model for COVID-19 patients","volume":"2","author":"Yan","journal-title":"Nat Mach Intell"},{"issue":"4","key":"R22","doi-asserted-by":"publisher","first-page":"844","DOI":"10.1111\/jth.14768","article-title":"Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia","volume":"18","author":"Tang","journal-title":"J Thromb Haemost"},{"issue":"9","key":"R23","doi-asserted-by":"publisher","first-page":"2128","DOI":"10.1016\/j.cgh.2020.04.040","article-title":"ACE2 expression in pancreas may cause pancreatic damage after SARS-CoV-2 infection","volume":"18","author":"Liu","journal-title":"Clin Gastroenterol Hepatol"},{"issue":"1","key":"R24","doi-asserted-by":"publisher","DOI":"10.1186\/s13054-020-03025-y","article-title":"Thrombotic and hemorrhagic events in critically ill COVID-19 patients: a French monocenter retrospective study","volume":"24","author":"Fraiss\u00e9","journal-title":"Crit Care"},{"key":"R25","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2019.02204","article-title":"The era of thromboinflammation: platelets are dynamic sensors and effector cells during infectious diseases","volume":"10","author":"Guo","journal-title":"Front Immunol"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2016.00236","article-title":"The emerging role of NETs in venous thrombosis and immunothrombosis","volume":"7","author":"Kimball","journal-title":"Front Immunol"},{"issue":"5","key":"R27","doi-asserted-by":"publisher","first-page":"506","DOI":"10.1097\/SHK.0000000000000645","article-title":"Gastrointestinal leakage detected by serum (1\u21923)-\u03b2-D-glucan in mouse models and a pilot study in patients with sepsis","volume":"46","author":"Leelahavanichkul","journal-title":"Shock"},{"issue":"7","key":"R28","doi-asserted-by":"publisher","DOI":"10.3390\/cells11071103","article-title":"Neutrophil extracellular traps in severe SARS-CoV-2 infection: a possible impact of LPS and (1\u21923)-\u03b2-D-glucan in blood from gut translocation","volume":"11","author":"Saithong","journal-title":"Cells"},{"issue":"5","key":"R29","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1016\/j.healun.2020.03.012","article-title":"COVID-19 illness in native and immunosuppressed states: a clinical-therapeutic staging proposal","volume":"39","author":"Siddiqi","journal-title":"J Heart Lung Transplant"},{"issue":"18","key":"R30","doi-asserted-by":"publisher","first-page":"1757","DOI":"10.1056\/NEJMcp2009249","article-title":"Mild or Moderate COVID-19","volume":"383","author":"Gandhi","journal-title":"N Engl J Med"},{"issue":"1","key":"R31","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1038\/s41581-022-00642-4","article-title":"Therapeutic advances in COVID-19","volume":"19","author":"Murakami","journal-title":"Nat Rev Nephrol"},{"issue":"5","key":"R32","doi-asserted-by":"publisher","first-page":"1094","DOI":"10.1111\/jth.14817","article-title":"Anticoagulant treatment is associated with decreased mortality in severe coronavirus disease 2019 patients with coagulopathy","volume":"18","author":"Tang","journal-title":"J Thromb Haemost"}],"container-title":["JMIR Formative Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/formative.jmir.org\/2025\/1\/e65585","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:29:31Z","timestamp":1785536971000},"score":1,"resource":{"primary":{"URL":"https:\/\/formative.jmir.org\/2025\/1\/e65585"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,7]]},"references-count":32,"URL":"https:\/\/doi.org\/10.2196\/65585","relation":{},"ISSN":["2561-326X"],"issn-type":[{"value":"2561-326X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,7]]},"article-number":"v9i3e65585"}}