{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T18:11:15Z","timestamp":1789755075584,"version":"4.0.1"},"reference-count":23,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T00:00:00Z","timestamp":1789689600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["VE 1274\/1-2"],"award-info":[{"award-number":["VE 1274\/1-2"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Chronic subdural hematoma (cSDH) recurrence requiring reoperation occurs in 5\u201333% of cases. Predicting recurrence could enable risk-stratified surveillance, reducing imaging in low-risk patients while maintaining monitoring for high-risk individuals. We evaluated whether machine learning could achieve clinically actionable recurrence prediction using routinely available variables.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This retrospective single-center study included 564 consecutive patients undergoing surgical cSDH evacuation (2015\u20132023), randomly divided into training (75%, n\u2009=\u2009422) and test (25%, n\u2009=\u2009142) sets. We developed and compared three models, regularized logistic regression, Random Forest, and XGBoost, using 31 predictor variables. Model development and tuning used 10-fold cross-validation on the training set; the best model was evaluated on the held-out test set. The primary outcome was postoperative recurrence requiring reoperation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Postoperative recurrence occurred in 170 patients (30.1%). In the training set, XGBoost achieved the highest cross-validated ROC AUC (0.713, SE\u2009=\u20090.024), matching Random Forest and outperforming logistic regression (0.686). Hematoma volume, coagulation parameters, and disease severity markers (ICU admission, GCS) were the most influential predictors, though effect sizes remained modest. On the test set, the final XGBoost model achieved ROC AUC 0.688 (95% CI 0.590\u20130.772), with satisfactory average calibration but overconfident individual-level risk estimates (calibration slope 0.615). At the clinically relevant 90% sensitivity threshold, specificity was only 30.3%, allowing potential imaging reduction in roughly one-third of non-recurrence patients. Consistency between training and test performance indicated these limitations reflect predictor information content rather than overfitting.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>In this single-center cohort, machine learning models using routinely available clinical and radiographic variables did not achieve clinically actionable risk stratification for cSDH recurrence under internal validation, with discriminative capacity insufficient to identify a low-risk subgroup suitable for de-escalated surveillance. These findings suggest recurrence is driven by factors not captured in standard clinical assessment, supporting uniform or symptom-driven imaging strategies over risk-stratified approaches.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0346756","type":"journal-article","created":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T17:51:13Z","timestamp":1789753873000},"page":"e0346756","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Classification of recurrence status after surgical treatment of chronic subdural hemorrhage \u2013 A machine learning approach"],"prefix":"10.1371","volume":"21","author":[{"given":"Hussam","family":"Hamou","sequence":"first","affiliation":[{"name":"RWTH Aachen University Hospital","place":["Aachen, Germany"],"department":["Department of Neurosurgery"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Julius","family":"Kernbach","sequence":"additional","affiliation":[{"name":"RWTH Aachen University Hospital","place":["Aachen, Germany"],"department":["Department of Neurosurgery"]},{"name":"University Hospital Heidelberg","place":["Heidelberg, Germany"],"department":["Department of Neuroradiology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hani","family":"Ridwan","sequence":"additional","affiliation":[{"name":"RWTH Aachen University","place":["Aachen, Germany"],"department":["Department of Diagnostic and Interventional Neuroradiology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kimberley","family":"Fay-Rodrian","sequence":"additional","affiliation":[{"name":"RWTH Aachen University Hospital","place":["Aachen, Germany"],"department":["Department of Neurosurgery"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8258-9346","authenticated-orcid":true,"given":"Hans","family":"Clusmann","sequence":"additional","affiliation":[{"name":"RWTH Aachen University Hospital","place":["Aachen, Germany"],"department":["Department of Neurosurgery"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anke","family":"Hoellig","sequence":"additional","affiliation":[{"name":"RWTH Aachen University Hospital","place":["Aachen, Germany"],"department":["Department of Neurosurgery"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3648-6842","authenticated-orcid":true,"given":"Michael","family":"Veldeman","sequence":"additional","affiliation":[{"name":"RWTH Aachen University Hospital","place":["Aachen, Germany"],"department":["Department of Neurosurgery"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,18]]},"reference":[{"issue":"9","key":"pone.0346756.ref001","doi-asserted-by":"crossref","first-page":"2033","DOI":"10.1007\/s00701-020-04398-3","article-title":"Chronic subdural hematoma-incidence, complications, and financial impact","volume":"162","author":"M Rauhala","year":"2020","journal-title":"Acta Neurochir (Wien)"},{"issue":"12","key":"pone.0346756.ref002","first-page":"208","article-title":"Chronic Subdural Hematoma","volume":"119","author":"HA Hamou","year":"2022","journal-title":"Dtsch Arztebl Int"},{"issue":"1","key":"pone.0346756.ref003","doi-asserted-by":"crossref","first-page":"222","DOI":"10.3171\/2016.9.JNS16623","article-title":"Present epidemiology of chronic subdural hematoma in Japan: analysis of 63,358 cases recorded in a national administrative database","volume":"128","author":"H Toi","year":"2018","journal-title":"J Neurosurg"},{"key":"pone.0346756.ref004","doi-asserted-by":"crossref","first-page":"1206996","DOI":"10.3389\/fneur.2023.1206996","article-title":"Incidence, therapy, and outcome in the management of chronic subdural hematoma in Switzerland: a population-based multicenter cohort study","volume":"14","author":"A El Rahal","year":"2023","journal-title":"Front Neurol"},{"issue":"6","key":"pone.0346756.ref005","doi-asserted-by":"crossref","first-page":"375","DOI":"10.1093\/qjmed\/hcw231","article-title":"Increasing incidence of chronic subdural haematoma in the elderly","volume":"110","author":"V Adhiyaman","year":"2017","journal-title":"QJM"},{"issue":"1","key":"pone.0346756.ref006","doi-asserted-by":"crossref","first-page":"108","DOI":"10.1186\/s12974-017-0881-y","article-title":"Pathophysiology of chronic subdural haematoma: inflammation, angiogenesis and implications for pharmacotherapy","volume":"14","author":"E Edlmann","year":"2017","journal-title":"J Neuroinflammation"},{"issue":"3","key":"pone.0346756.ref007","doi-asserted-by":"crossref","first-page":"1353","DOI":"10.1007\/s11357-022-00570-y","article-title":"The pathophysiology of chronic subdural hematoma revisited: emphasis on aging processes as key factor","volume":"44","author":"R Weigel","year":"2022","journal-title":"Geroscience"},{"issue":"10","key":"pone.0346756.ref008","doi-asserted-by":"crossref","first-page":"570","DOI":"10.1038\/nrneurol.2014.163","article-title":"Chronic subdural haematoma: modern management and emerging therapies","volume":"10","author":"AG Kolias","year":"2014","journal-title":"Nat Rev Neurol"},{"key":"pone.0346756.ref009","doi-asserted-by":"crossref","first-page":"101234","DOI":"10.1016\/j.eclinm.2021.101234","article-title":"Factors correlated with the postoperative recurrence of chronic subdural hematoma: An umbrella study of systematic reviews and meta-analyses","volume":"43","author":"F Zhu","year":"2022","journal-title":"eClinicalMedicine"},{"issue":"4","key":"pone.0346756.ref010","doi-asserted-by":"crossref","first-page":"2777","DOI":"10.1007\/s10143-022-01790-8","article-title":"Risk factors of recurrence in chronic subdural hematoma and a proposed extended classification of internal architecture as a predictor of recurrence","volume":"45","author":"H Hamou","year":"2022","journal-title":"Neurosurg Rev"},{"issue":"12","key":"pone.0346756.ref011","doi-asserted-by":"crossref","first-page":"1186","DOI":"10.1056\/NEJMc1812507","article-title":"Follow-up Computed Tomography after Evacuation of Chronic Subdural Hematoma","volume":"380","author":"P Schucht","year":"2019","journal-title":"N Engl J Med"},{"issue":"5","key":"pone.0346756.ref012","doi-asserted-by":"crossref","first-page":"885","DOI":"10.1007\/s00701-019-03858-9","article-title":"The Danish chronic subdural hematoma study-predicting recurrence of chronic subdural hematoma","volume":"161","author":"NC Andersen-Ranberg","year":"2019","journal-title":"Acta Neurochir (Wien)"},{"key":"pone.0346756.ref013","doi-asserted-by":"crossref","first-page":"104286","DOI":"10.1016\/j.bas.2025.104286","article-title":"Recurrence prediction in chronic subdural hematomas: a risk stratification score based on 118 consecutive patients","volume":"5","author":"FMC Lioi","year":"2025","journal-title":"Brain Spine"},{"issue":"4","key":"pone.0346756.ref014","article-title":"Predictors of Recurrence of Chronic Subdural Hematoma Among Adult Filipinos After Surgery: Developing a Preliminary Model for Prognosis","volume":"17","author":"CMC Mercado","year":"2025","journal-title":"Cureus"},{"issue":"5","key":"pone.0346756.ref015","doi-asserted-by":"crossref","first-page":"1143","DOI":"10.3171\/2017.6.JNS17130","article-title":"Factors predicting reoperation of chronic subdural hematoma following primary surgical evacuation","volume":"129","author":"R Motiei-Langroudi","year":"2018","journal-title":"J Neurosurg"},{"key":"pone.0346756.ref016","doi-asserted-by":"crossref","first-page":"33","DOI":"10.1007\/978-3-030-85292-4_5","article-title":"Foundations of Machine Learning-Based Clinical Prediction Modeling: Part IV-A Practical Approach to Binary Classification Problems","volume":"134","author":"VE Staartjes","year":"2022","journal-title":"Acta Neurochir Suppl"},{"key":"pone.0346756.ref017","doi-asserted-by":"crossref","first-page":"977329","DOI":"10.3389\/fneur.2022.977329","article-title":"Seizure after surgical treatment of chronic subdural hematoma-Associated factors and effect on outcome","volume":"13","author":"H Hamou","year":"2022","journal-title":"Front Neurol"},{"issue":"16","key":"pone.0346756.ref018","doi-asserted-by":"crossref","first-page":"4591","DOI":"10.3390\/jcm13164591","article-title":"Effects of Angiotensin-Converting Enzyme Inhibition on the Recurrence and Internal Structure of Chronic Subdural Hematomas","volume":"13","author":"M Veldeman","year":"2024","journal-title":"J Clin Med"},{"issue":"12","key":"pone.0346756.ref019","article-title":"Guidelines for Developing and Reporting Machine Learning Predictive Models in Biomedical Research: A Multidisciplinary View","volume":"18","author":"W Luo","year":"2016","journal-title":"J Med Internet Res"},{"issue":"4","key":"pone.0346756.ref020","doi-asserted-by":"crossref","first-page":"1077","DOI":"10.3171\/2022.8.JNS22888","article-title":"Automated detection and analysis of subdural hematomas using a machine learning algorithm","volume":"138","author":"M Colasurdo","year":"2022","journal-title":"J Neurosurg"},{"key":"pone.0346756.ref021","doi-asserted-by":"crossref","DOI":"10.2196\/54944","article-title":"Combining Clinical-Radiomics Features With Machine Learning Methods for Building Models to Predict Postoperative Recurrence in Patients With Chronic Subdural Hematoma: Retrospective Cohort Study","volume":"26","author":"C Fang","year":"2024","journal-title":"J Med Internet Res"},{"issue":"19","key":"pone.0346756.ref022","doi-asserted-by":"crossref","first-page":"4230","DOI":"10.1002\/sim.9025","article-title":"Minimum sample size for external validation of a clinical prediction model with a binary outcome","volume":"40","author":"RD Riley","year":"2021","journal-title":"Stat Med"},{"issue":"10","key":"pone.0346756.ref023","doi-asserted-by":"crossref","first-page":"2719","DOI":"10.1007\/s00701-022-05216-8","article-title":"External validation of prognostic models predicting outcome after chronic subdural hematoma","volume":"164","author":"DC Holl","year":"2022","journal-title":"Acta Neurochir (Wien)"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346756","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T17:51:17Z","timestamp":1789753877000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346756"}},"subtitle":[],"editor":[{"given":"Michael C","family":"Burger","sequence":"first","affiliation":[{"name":"Goethe University Hospital Frankfurt","place":["GERMANY"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,18]]},"references-count":23,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,18]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0346756","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,18]]}}}