{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T04:30:57Z","timestamp":1786681857298,"version":"build-2736575974"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T00:00:00Z","timestamp":1786665600000},"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. Neurol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Deep vein thrombosis (DVT) is a common complication of acute ischemic stroke (AIS) and may worsen clinical outcomes, yet reliable tools for early risk stratification remain limited. We aimed to develop and internally evaluate machine learning models for predicting in-hospital DVT in patients with AIS.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We conducted a secondary analysis of a publicly available multicenter retrospective dataset including 21,459 patients with AIS. The primary outcome was imaging-confirmed in-hospital DVT. Participants were stratified according to DVT status and randomly divided into a training set (70%) and a held-out test set (30%). Feature selection was performed using least absolute shrinkage and selection operator regression, the Boruta algorithm, variance inflation factor assessment, and clinical judgment. Eight machine learning algorithms were trained using a 19-variable full predictor set and an 8-variable simplified predictor set. Hyperparameters were optimized using repeated 5-fold cross-validation with 2 repeats. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), area under the precision-recall curve (AUPRC), Brier score, calibration, decision curve analysis, and additional classification metrics. Sensitivity analyses excluded D-dimer and used within-fold synthetic minority oversampling.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Among 21,459 patients, 1,324 (6.17%) developed in-hospital DVT. In the full predictor-set analysis, RANGER achieved the highest AUC in the held-out test set (0.976). Among models using the simplified predictor set, XGBoost achieved the highest AUC (0.917) and sensitivity (0.852), whereas SVM demonstrated the most favorable overall performance profile, with the highest AUPRC (0.605), lowest Brier score (0.038), highest positive predictive value (0.440), and highest\n                      <jats:italic>F<\/jats:italic>\n                      1-score (0.549). D-dimer was the most influential predictor. Model performance was attenuated after exclusion of D-dimer, and the SMOTE analysis showed slightly lower discrimination and greater calibration discrepancies. All 8 simplified predictor-set models were incorporated into an online prediction platform.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Machine learning models demonstrated favorable performance for predicting in-hospital DVT after AIS. Among models using the simplified 8-variable predictor set, SVM showed the most favorable overall performance, whereas XGBoost prioritized sensitivity. Independent external validation and prospective clinical-impact assessment are required before routine clinical implementation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fneur.2026.1847449","type":"journal-article","created":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T04:23:08Z","timestamp":1786681388000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Machine learning\u2013based prediction of in-hospital deep vein thrombosis in patients with acute ischemic stroke: a multicenter study"],"prefix":"10.3389","volume":"17","author":[{"given":"Tieshi","family":"Zhu","sequence":"first","affiliation":[{"name":"Department of Neurology, Zhanjiang Central Hospital, Guangdong Medical University","place":["Zhanjiang, Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Runzhui","family":"Lin","sequence":"additional","affiliation":[{"name":"Department of Hepatobiliary, Pancreatic, and Splenic Surgery, Second Affiliated Hospital of Shantou University Medical College","place":["Shantou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Le","family":"Zhao","sequence":"additional","affiliation":[{"name":"Department of Medical Affairs, Central Hospital of Guangdong Provincial Nongken","place":["Zhanjiang, Guangdong, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"He","family":"Zhu","sequence":"additional","affiliation":[{"name":"Department of Clinical Research Institute, Zhanjiang Central Hospital, Guangdong Medical University","place":["Zhanjiang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,14]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1159\/000549818","article-title":"Global burden of ischemic stroke attributable to high sugar-sweetened beverages consumption from 1990 to 2021 and projections to 2045","author":"Xue","year":"2026","journal-title":"Neuroepidemiology"},{"key":"ref2","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","year":"2024","journal-title":"Lancet Neurol"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"108543","DOI":"10.1016\/j.jstrokecerebrovasdis.2026.108543","article-title":"Global and regional burden of ischemic stroke attributable to smoking and high alcohol use from 1990 to 2021, with projections to 2050","volume":"35","author":"Guo","year":"2026","journal-title":"J Stroke Cerebrovasc Dis"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"e1282","DOI":"10.1212\/wnl.0000000000206749","article-title":"Age- and sex-specific trends in medical complications after acute ischemic stroke in the United States","volume":"100","author":"Ahmed","year":"2023","journal-title":"Neurology"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"e26","DOI":"10.1016\/j.wneu.2020.04.038","article-title":"Short-term and long-term safety and efficacy of treatment of acute ischemic stroke with low-molecular-weight heparin: meta-analysis of 19 randomized controlled trials","volume":"141","author":"Ye","year":"2020","journal-title":"World Neurosurg"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"e000372","DOI":"10.1161\/jaha.113.000372","article-title":"Pulmonary embolism in ischemic stroke: clinical presentation, risk factors, and outcome","volume":"2","author":"Pongmoragot","year":"2013","journal-title":"J Am Heart Assoc"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"104623","DOI":"10.1016\/j.ijnurstu.2023.104623","article-title":"Risk prediction models for deep venous thrombosis in patients with acute stroke: a systematic review and meta-analysis","volume":"149","author":"Fu","year":"2024","journal-title":"Int J Nurs Stud"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"432","DOI":"10.1111\/j.1747-4949.2009.00348.x","article-title":"Primary end-point times, functional outcome and adverse event profile after acute ischaemic stroke","volume":"4","author":"Ali","year":"2009","journal-title":"Int J Stroke"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"1516","DOI":"10.1161\/01.str.27.9.1516","article-title":"Deep vein thrombosis in stroke. The use of plasma D-dimer level as a screening test in the rehabilitation setting","volume":"27","author":"Harvey","year":"1996","journal-title":"Stroke"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"e316","DOI":"10.1161\/str.0000000000000513","article-title":"2026 guideline for the early management of patients with acute ischemic stroke: a guideline from the American Heart Association\/American Stroke Association","volume":"57","author":"Prabhakaran","year":"2026","journal-title":"Stroke"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1227","DOI":"10.1001\/jamaneurol.2025.3662","article-title":"Optimal antithrombotics for ischemic stroke and concurrent atrial fibrillation and atherosclerosis: a randomized clinical trial","volume":"82","author":"Okazaki","year":"2025","journal-title":"JAMA Neurol"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"e31585","DOI":"10.1097\/md.0000000000031585","article-title":"A nomogram for individualized prediction of lower extremity deep venous thrombosis in stroke patients: a retrospective study","volume":"101","author":"Liu","year":"2022","journal-title":"Medicine"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"105683","DOI":"10.1016\/j.jstrokecerebrovasdis.2021.105683","article-title":"Development and validation of a nomogram for lower extremity deep venous thrombosis in patients after acute stroke","volume":"30","author":"Pan","year":"2021","journal-title":"J Stroke Cerebrovasc Dis"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.jocn.2024.04.022","article-title":"Development and validation of a risk prediction model for aspiration in patients with acute ischemic stroke","volume":"124","author":"Wang","year":"2024","journal-title":"J Clin Neurosci"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"140","DOI":"10.1186\/s12877-021-02088-y","article-title":"Individualized predictions of early isolated distal deep vein thrombosis in patients with acute ischemic stroke: a retrospective study","volume":"21","author":"Cheng","year":"2021","journal-title":"BMC Geriatr"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"e181404","DOI":"10.1001\/jamanetworkopen.2018.1404","article-title":"Machine learning for prediction in electronic health data","volume":"1","author":"Rose","year":"2018","journal-title":"JAMA Netw Open"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1038\/s41746-022-00712-8","article-title":"Multimodal machine learning in precision health: a scoping review","volume":"5","author":"Kline","year":"2022","journal-title":"NPJ Digit Med"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"RP98759","DOI":"10.7554\/eLife.98759","article-title":"Predictive models for secondary epilepsy in patients with acute ischemic stroke within one year","volume":"13","author":"Liu","year":"2024","journal-title":"eLife"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"403","DOI":"10.1111\/cns.12227","article-title":"Risk assessment of deep-vein thrombosis after acute stroke: a prospective study using clinical factors","volume":"20","author":"Liu","year":"2014","journal-title":"CNS Neurosci Ther"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"10760296231171603","DOI":"10.1177\/10760296231171603","article-title":"Nomogram prediction for lower extremity deep vein thrombosis in acute ischemic stroke patients receiving thrombolytic therapy","volume":"29","author":"Wang","year":"2023","journal-title":"Clin Appl Thromb Hemost"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1506959","DOI":"10.3389\/fneur.2025.1506959","article-title":"Developing a predictive model for lower extremity deep vein thrombosis in acute ischemic stroke using a nomogram","volume":"16","author":"Chen","year":"2025","journal-title":"Front Neurol"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"e40754","DOI":"10.1097\/md.0000000000040754","article-title":"Construction and validation of a prediction model for acute ischemic stroke patients with concomitant deep vein thrombosis","volume":"103","author":"Xu","year":"2024","journal-title":"Medicine"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"e0302676","DOI":"10.1371\/journal.pone.0302676","article-title":"Leveraging machine learning for enhanced and interpretable risk prediction of venous thromboembolism in acute ischemic stroke care","volume":"20","author":"Jiang","year":"2025","journal-title":"PLoS One"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"516","DOI":"10.5694\/mja2.50201","article-title":"Deep vein thrombosis: update on diagnosis and management","volume":"210","author":"Kruger","year":"2019","journal-title":"Med J Aust"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"2173","DOI":"10.1111\/jth.15813","article-title":"Risk factors and predictors for venous thromboembolism in people with ischemic stroke: a systematic review","volume":"20","author":"T\u00f8ndel","year":"2022","journal-title":"J Thromb Haemost"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"644","DOI":"10.1016\/j.jvs.2004.07.004","article-title":"C-reactive protein and red cell aggregation correlate with late venous function after acute deep venous thrombosis","volume":"40","author":"Krieger","year":"2004","journal-title":"J Vasc Surg"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"282","DOI":"10.1016\/j.avsg.2018.06.013","article-title":"Correlation of Interleukin-18 and high-sensitivity C-reactive protein with perioperative deep vein thrombosis in patients with ankle fracture","volume":"54","author":"Du","year":"2019","journal-title":"Ann Vasc Surg"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1002652","DOI":"10.3389\/fimmu.2022.1002652","article-title":"C-reactive protein, immunothrombosis and venous thromboembolism","volume":"13","author":"Dix","year":"2022","journal-title":"Front Immunol"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"2823","DOI":"10.1016\/j.jtha.2024.06.018","article-title":"Risk of venous thromboembolism and arterial events in patients with hypoalbuminemia: a comprehensive meta-analysis of more than 2 million patients","volume":"22","author":"Valeriani","year":"2024","journal-title":"J Thromb Haemost"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"620","DOI":"10.3340\/jkns.2016.1010.009","article-title":"Elevated blood urea nitrogen\/creatinine ratio is associated with venous thromboembolism in patients with acute ischemic stroke","volume":"60","author":"Kim","year":"2017","journal-title":"J Korean Neurosurg Soc"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1007\/s11239-020-02181-w","article-title":"Dvt incidence and risk factors in critically ill patients with COVID-19","volume":"51","author":"Chen","year":"2021","journal-title":"J Thromb Thrombolysis"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"100555","DOI":"10.1016\/j.jacadv.2023.100555","article-title":"Venous thromboembolism in patients with atrial fibrillation: a systematic review and meta-analysis of 4,170,027 patients","volume":"2","author":"Pastori","year":"2023","journal-title":"JACC Adv"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"5293","DOI":"10.1111\/jocn.17786","article-title":"Nomogram for deep vein thrombosis prediction post-endovascular thrombectomy in acute ischemic stroke: a retrospective multicenter observational study","volume":"34","author":"Han","year":"2025","journal-title":"J Clin Nurs"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"416","DOI":"10.1001\/jama.2025.21561","article-title":"Age-adjusted D-dimer cutoff levels to rule out deep vein thrombosis","volume":"335","author":"Le Gal","year":"2026","journal-title":"JAMA"}],"container-title":["Frontiers in Neurology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1847449\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T04:23:11Z","timestamp":1786681391000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2026.1847449\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,14]]},"references-count":34,"alternative-id":["10.3389\/fneur.2026.1847449"],"URL":"https:\/\/doi.org\/10.3389\/fneur.2026.1847449","relation":{},"ISSN":["1664-2295"],"issn-type":[{"value":"1664-2295","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,14]]},"article-number":"1847449"}}