{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T20:33:30Z","timestamp":1786394010661,"version":"build-2736575974"},"reference-count":49,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n The medical burden caused by stroke is increasingly severe, and a small minority of high-cost patients consume the majority of medical expenditures. Therefore, revealing the formation mechanisms of this population and exploring a scientific cost-risk stratification system are crucial for improving the quality of care and achieving the optimal allocation of medical resources.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to construct a comorbidity network for patients with stroke using standardized front-page medical record data, extract network features that reflect complex disease interactions, and develop identification models in combination with machine learning algorithms. The study focused on building a core model integrating variables from the near-discharge stage for stratifying the risk of high hospitalization costs in patients at the near-discharge stage. In addition, an early prediction model was developed using only data available at admission.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We conducted a retrospective study, collecting the hospital discharge data of inpatients with stroke from a tertiary hospital in Northeast China between 2021 and 2023. The data from 2021 to 2022 were used to construct a network and extract features to capture the potential relationship between diseases and high costs. Using the 2023 data partitioned into training and testing sets, we developed 5 models to identify inpatients with stroke who incurred high hospitalization costs and compared their performance when input with different features. In addition, the Shapley Additive Explanations interpretability method was adopted to explain the global and local contributions of the model features.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n The inclusion of network features significantly improved the model\u2019s performance, among which Extreme Gradient Boosting performed the best. The global feature importance showed that network features occupied a major proportion. The results of the Shapley Additive Explanations interaction analysis indicated potential phased changes in patient resource consumption. However, the overall performance of the early identification model constructed solely from admission data was subject to clear limitations.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n This study developed an integrated framework combining comorbidity network analysis with machine learning, which significantly improved the accuracy of identifying inpatients with stroke at high risk of incurring excessive hospitalization costs. The core model demonstrated good performance in risk stratification during the near-discharge stage, showing potential for application in the formulation of risk management strategies and the optimization of health care resource allocation. It also laid the foundation for the subsequent development of more accurate early identification models.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/93680","type":"journal-article","created":{"date-parts":[[2026,7,8]],"date-time":"2026-07-08T15:50:06Z","timestamp":1783525806000},"page":"e93680-e93680","source":"Crossref","is-referenced-by-count":0,"title":["Network Analysis\u2013Driven Machine Learning Model for Identifying High-Cost Stroke Inpatients Using Hospital Discharge Data: Retrospective Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-1439-7423","authenticated-orcid":false,"given":"Haohui","family":"Shen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2210-6311","authenticated-orcid":false,"given":"Yilong","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-8096-3246","authenticated-orcid":false,"given":"Mengge","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-1751-8484","authenticated-orcid":false,"given":"Jingyi","family":"Xiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7013-458X","authenticated-orcid":false,"given":"Runan","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-3674-9249","authenticated-orcid":false,"given":"Pin","family":"Yao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,8,10]]},"reference":[{"issue":"10","key":"R1","doi-asserted-by":"publisher","first-page":"973","DOI":"10.1016\/S1474-4422(24)00369-7","article-title":"Global, regional, and national burden of stroke and its risk factors, 1990\u20132021: a systematic analysis for the Global Burden of Disease Study 2021","volume":"23","author":"Feigin","journal-title":"Lancet Neurol"},{"issue":"2","key":"R2","doi-asserted-by":"publisher","first-page":"132","DOI":"10.1177\/17474930241308142","article-title":"World Stroke Organization: Global Stroke Fact Sheet 2025","volume":"20","author":"Feigin","journal-title":"Int J Stroke"},{"issue":"3","key":"R3","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2023-080634","article-title":"Estimating the economic burden of stroke in China: a cost-of-illness study","volume":"14","author":"Ma","journal-title":"BMJ Open"},{"issue":"4","key":"R4","first-page":"41","volume":"9","author":"Garfinkel","journal-title":"Health Care Financ Rev"},{"issue":"6","key":"R5","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0217353","article-title":"A challenge to all. A primer on inter-country differences of high-need, high-cost patients","volume":"14","author":"Tanke","journal-title":"PLOS ONE"},{"issue":"4","key":"R6","doi-asserted-by":"publisher","first-page":"346","DOI":"10.1136\/archdischild-2021-321654","article-title":"Categorising high-cost high-need children and young people","volume":"107","author":"Punjabi","journal-title":"Arch Dis Child"},{"issue":"11","key":"R7","doi-asserted-by":"publisher","first-page":"5761","DOI":"10.3390\/su13115761","article-title":"Concentration and persistence of healthcare spending: evidence from China","volume":"13","author":"Chen","journal-title":"Sustainability"},{"issue":"9","key":"R8","doi-asserted-by":"publisher","first-page":"807","DOI":"10.1056\/NEJMp1612779","article-title":"Focusing on high-cost patients - the key to addressing high costs?","volume":"376","author":"McWilliams","journal-title":"N Engl J Med"},{"issue":"5","key":"R9","doi-asserted-by":"publisher","first-page":"1002","DOI":"10.1111\/jgs.15257","article-title":"Managing high-cost healthcare users: the international search for effective evidence-supported strategies","volume":"66","author":"Lee","journal-title":"J Am Geriatr Soc"},{"issue":"9","key":"R10","doi-asserted-by":"publisher","first-page":"1815","DOI":"10.1007\/s11606-019-05082-8","article-title":"Effect of intensive interdisciplinary transitional care for high-need, high-cost patients on quality, outcomes, and costs: a quasi-experimental study","volume":"34","author":"Bailey","journal-title":"J Gen Intern Med"},{"issue":"2","key":"R11","doi-asserted-by":"publisher","first-page":"100","DOI":"10.1089\/pop.2022.0271","article-title":"Differential impact of a plan-led standardized complex care management intervention on subgroups of high-cost high-need Medicaid patients","volume":"26","author":"Quinton","journal-title":"Popul Health Manag"},{"issue":"10","key":"R12","doi-asserted-by":"publisher","first-page":"909","DOI":"10.1056\/NEJMp1608511","article-title":"Caring for high-need, high-cost patients - an urgent priority","volume":"375","author":"Blumenthal","journal-title":"N Engl J Med"},{"issue":"6","key":"R13","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2021-058539","article-title":"Unpacking complex interventions that manage care for high-need, high-cost patients: a realist review","volume":"12","author":"Chang","journal-title":"BMJ Open"},{"issue":"9","key":"R14","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2018-023113","article-title":"Systematic review of high-cost patients\u2019 characteristics and healthcare utilisation","volume":"8","author":"Wammes","journal-title":"BMJ Open"},{"issue":"1","key":"R15","doi-asserted-by":"publisher","DOI":"10.1186\/s12888-024-06358-y","article-title":"Machine-learning-based cost prediction models for inpatients with mental disorders in China","volume":"25","author":"Ma","journal-title":"BMC Psychiatry"},{"issue":"1","key":"R16","doi-asserted-by":"publisher","DOI":"10.1186\/s12913-020-05936-6","article-title":"Identifying and understanding determinants of high healthcare costs for breast cancer: a quantile regression machine learning approach","volume":"20","author":"Hu","journal-title":"BMC Health Serv Res"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-020-00354-8","article-title":"Machine-learning-based prediction models for high-need high-cost patients using nationwide clinical and claims data","volume":"3","author":"Osawa","journal-title":"NPJ Digit Med"},{"issue":"9","key":"R18","doi-asserted-by":"publisher","first-page":"987","DOI":"10.1177\/1747493016660107","article-title":"Changes in the prevalence of hospitalization and comorbidity in US adults with stroke: a three decade cross-sectional and birth cohort analysis","volume":"11","author":"Liu","journal-title":"Int J Stroke"},{"issue":"6","key":"R19","doi-asserted-by":"publisher","first-page":"1429","DOI":"10.1007\/s00415-019-09278-0","article-title":"Heterogeneity in costs and prognosis for acute ischemic stroke treatment by comorbidities","volume":"266","author":"Han","journal-title":"J Neurol"},{"issue":"2","key":"R20","doi-asserted-by":"publisher","first-page":"242","DOI":"10.1111\/jth.13908","article-title":"Comorbidity and the increased mortality after hospitalization for stroke: a population-based cohort study","volume":"16","author":"Corraini","journal-title":"J Thromb Haemost"},{"issue":"9","key":"R21","doi-asserted-by":"publisher","first-page":"1999","DOI":"10.1017\/S0950268815003271","article-title":"Administrative data is as good as medical chart review for comorbidity ascertainment in patients with infections in Singapore","volume":"144","author":"Hwang","journal-title":"Epidemiol Infect"},{"issue":"3","key":"R22","doi-asserted-by":"publisher","first-page":"396","DOI":"10.1016\/j.jos.2020.04.005","article-title":"Predicting in-hospital mortality for dementia patients after hip fracture surgery - A comparison between the Charlson Comorbidity Index (CCI) and the Elixhauser Comorbidity Index","volume":"26","author":"Tang","journal-title":"J Orthop Sci"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-33251-9","article-title":"Interpretable machine learning based on the Charlson comorbidity index predicts 28-day mortality in acute hypercapnic respiratory failure","volume":"16","author":"Lu","journal-title":"Sci Rep"},{"issue":"5","key":"R24","doi-asserted-by":"publisher","first-page":"373","DOI":"10.1016\/0021-9681(87)90171-8","article-title":"A new method of classifying prognostic comorbidity in longitudinal studies: development and validation","volume":"40","author":"Charlson","journal-title":"J Chronic Dis"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1097\/00005650-199801000-00004","article-title":"Comorbidity measures for use with administrative data","volume":"36","author":"Elixhauser","journal-title":"Med Care"},{"issue":"1","key":"R26","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-024-02286-3","article-title":"Choices of measures of association affect the visualisation and composition of the multimorbidity networks","volume":"24","author":"Baneshi","journal-title":"BMC Med Res Methodol"},{"key":"R27","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2019.103206","article-title":"Disease networks and their contribution to disease understanding: a review of their evolution, techniques and data sources","volume":"94","author":"Garc\u00eda Del Valle","journal-title":"J Biomed Inform"},{"issue":"8","key":"R28","doi-asserted-by":"publisher","first-page":"1217","DOI":"10.1007\/s10389-021-01685-w","article-title":"Methods of analysing patterns of multimorbidity using network analysis: a scoping review","volume":"31","author":"Jones","journal-title":"J Public Health (Berl)"},{"key":"R29","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2020.08.044","article-title":"Predicting post-discharge self-harm incidents using disease comorbidity networks: a retrospective machine learning study","volume":"277","author":"Xu","journal-title":"J Affect Disord"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-022-01802-z","article-title":"Network analytics and machine learning for predicting length of stay in elderly patients with chronic diseases at point of admission","volume":"22","author":"Hu","journal-title":"BMC Med Inform Decis Mak"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2022.118541","article-title":"Early prediction of high-cost inpatients with ischemic heart disease using network analytics and machine learning","volume":"210","author":"Yang","journal-title":"Expert Syst Appl"},{"issue":"1","key":"R32","doi-asserted-by":"publisher","DOI":"10.1186\/s12933-025-02613-y","article-title":"Assessment of stress hyperglycemia ratio to predict all-cause mortality in patients with critical cerebrovascular disease: a retrospective cohort study from the MIMIC-IV database","volume":"24","author":"Chen","journal-title":"Cardiovasc Diabetol"},{"issue":"4","key":"R33","doi-asserted-by":"publisher","first-page":"355","DOI":"10.1001\/jama.2012.216476","article-title":"Diagnoses and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia","volume":"309","author":"Dharmarajan","journal-title":"JAMA"},{"issue":"1","key":"R34","doi-asserted-by":"publisher","DOI":"10.1186\/s12913-023-09193-1","article-title":"Comparison of 30-day planned and unplanned readmissions in a tertiary teaching hospital in China","volume":"23","author":"Zhang","journal-title":"BMC Health Serv Res"},{"issue":"6","key":"R35","doi-asserted-by":"publisher","first-page":"1970","DOI":"10.1109\/JBHI.2017.2783049","article-title":"Predicting high-cost patients at point of admission using network science","volume":"22","author":"Srinivasan","journal-title":"IEEE J Biomed Health Inform"},{"key":"R36","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyt.2017.00114","article-title":"Predicting future high-cost schizophrenia patients using high-dimensional administrative data","volume":"8","author":"Wang","journal-title":"Front Psychiatry"},{"issue":"2","key":"R37","doi-asserted-by":"publisher","first-page":"532","DOI":"10.1111\/j.1475-6773.2009.01080.x","article-title":"Using information on clinical conditions to predict high-cost patients","volume":"45","author":"Fleishman","journal-title":"Health Serv Res"},{"issue":"10","key":"R38","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-024-10884-2","article-title":"Handling imbalanced medical datasets: review of a decade of research","volume":"57","author":"Salmi","journal-title":"Artif Intell Rev"},{"key":"R39","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2025.103286","article-title":"Identification and validation of an explainable early-stage chronic kidney disease prediction model: a multicenter retrospective study","volume":"84","author":"He","journal-title":"EClinicalMedicine"},{"key":"R40","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2023.102409","article-title":"Identification and validation of an explainable prediction model of acute kidney injury with prognostic implications in critically ill children: a prospective multicenter cohort study","volume":"68","author":"Hu","journal-title":"EClinicalMedicine"},{"issue":"1","key":"R41","doi-asserted-by":"publisher","DOI":"10.1186\/s12933-025-02654-3","article-title":"Relationship between atherogenic index of plasma and length of stay in critically ill patients with atherosclerotic cardiovascular disease: a retrospective cohort study and predictive modeling based on machine learning","volume":"24","author":"Guo","journal-title":"Cardiovasc Diabetol"},{"issue":"1","key":"R42","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-99546-z","article-title":"Predicting high-need high-cost pediatric hospitalized patients in China based on machine learning methods","volume":"15","author":"Zhang","journal-title":"Sci Rep"},{"issue":"1","key":"R43","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0279540","article-title":"The application of machine learning to predict high-cost patients: a performance-comparison of different models using healthcare claims data","volume":"18","author":"Langenberger","journal-title":"PLoS ONE"},{"issue":"1","key":"R44","doi-asserted-by":"publisher","DOI":"10.1186\/s13561-023-00422-1","article-title":"Predicting high health-cost users among people with cardiovascular disease using machine learning and nationwide linked social administrative datasets","volume":"13","author":"Nghiem","journal-title":"Health Econ Rev"},{"issue":"2","key":"R45","doi-asserted-by":"publisher","first-page":"431","DOI":"10.1016\/j.mayocpiqo.2021.01.015","article-title":"Clinical and economic burden of stroke among young, midlife, and older adults in the United States, 2002-2017","volume":"5","author":"Khan","journal-title":"Mayo Clin Proc Innov Qual Outcomes"},{"key":"R46","doi-asserted-by":"publisher","DOI":"10.3389\/fphar.2020.596183","article-title":"Hospitalization expenditures and out-of-pocket expenses in patients with stroke in Northeast China, 2015-2017: a pooled cross-sectional study","volume":"11","author":"Ma","journal-title":"Front Pharmacol"},{"issue":"10","key":"R47","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2020-040437","article-title":"Utilisation of health services among urban patients who had an ischaemic stroke with different health insurance - a cross-sectional study in China","volume":"10","author":"Yang","journal-title":"BMJ Open"},{"key":"R48","unstructured":"Hoyt G Chatterjee N Battaglia F Basu P . Medical applications of graph convolutional networks using electronic health records: a survey. arXiv. Preprint posted online on Feb 13, 2025. [doi: 10.48550\/arXiv.2502.09781]"},{"key":"R49","unstructured":"Gardinazzi Y March RG Kalahasti S Ramirez AM Neri M Nguyen C et al. Characterization of diseases in temporal comorbidity networks. arXiv. Preprint posted online on Jun 29, 2025. [doi: 10.48550\/arXiv.2506.22136]"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T20:12:43Z","timestamp":1786392763000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e93680"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,10]]},"references-count":49,"URL":"https:\/\/doi.org\/10.2196\/93680","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,10]]},"article-number":"v14i10e93680"}}