{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,30]],"date-time":"2026-08-30T05:16:43Z","timestamp":1788067003462,"version":"build-2784847793"},"reference-count":57,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T00:00:00Z","timestamp":1750204800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002947","name":"Anhui Medical University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100002947","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurol."],"abstract":"<jats:sec><jats:title>Objective<\/jats:title><jats:p>Pulmonary infection (PI) remains a prevalent and severe complication in patients recovering from spontaneous deep subcortical intracerebral hemorrhage (deep SICH). Accurate prediction of PI risk is crucial for early intervention and optimized clinical management. The aim of this study was to develop a machine learning (ML) model for predicting PI risk in patients during the recovery phase of deep SICH and to investigate the contributions of individual risk factors through explainable artificial intelligence techniques.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We conducted a retrospective study involving 649 patients diagnosed with PI during the recovery phase of deep SICH between 2021 and 2023. The cohort was divided into a training set (70%, <jats:italic>n<\/jats:italic> = 454) and a testing set (30%, <jats:italic>n<\/jats:italic> = 195). Eight key clinical features were identified using the Boruta algorithm: mechanical ventilation, nasogastric feeding, tracheotomy, antibacterial drug use, hyperbaric oxygen therapy, procalcitonin levels, sedative drug use, and consciousness scores. Seven ML algorithms were employed to build predictive models, with performance evaluated based on the area under the receiver operating characteristic (AUC) curve, sensitivity, specificity, and accuracy. The best-performing model was selected, and SHAP (Shapley Additive Explanations) analysis was performed to interpret feature importance.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Among 649 patients with deep SICH, no significant baseline differences were found between the training (<jats:italic>n<\/jats:italic> = 454) and testing (<jats:italic>n<\/jats:italic> = 195) sets. The Boruta algorithm identified eight key predictors of pulmonary infection (PI). The random forest (RF) model achieved the highest AUCs: 0.994 (95% CI: 0.989\u20130.998) in training and 0.931 (95% CI: 0.899\u20130.963) in testing. DeLong tests showed RF significantly outperformed several models (DT, SVM, LightGBM), while performance differences with XGBoost (<jats:italic>p<\/jats:italic> = 0.95), KNN (<jats:italic>p<\/jats:italic> = 0.80), and LR (<jats:italic>p<\/jats:italic> = 0.22) were not significant. SHAP analysis revealed mechanical ventilation, nasogastric feeding, and tracheotomy as key risk factors, with hyperbaric oxygen therapy and higher consciousness scores showing protective effects.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>This study provides a high-performing and interpretable ML-based risk stratification tool for pulmonary infection in patients during the recovery phase of deep SICH. The integration of SHAP enhances clinical applicability by demystifying complex model outputs, thereby supporting individualized preventive strategies. These findings underscore the promise of explainable AI in advancing neurocritical care and call for prospective multicenter validation and real-time dynamic model adaptation in future research.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fneur.2025.1571755","type":"journal-article","created":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T05:46:46Z","timestamp":1750225606000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Construction of a risk prediction model for pulmonary infection in patients with spontaneous intracerebral hemorrhage during the recovery phase based on machine learning"],"prefix":"10.3389","volume":"16","author":[{"given":"Jixiang","family":"Xu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuan","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fumin","family":"Zhu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoxiao","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liang","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yinliang","family":"Qi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaomei","family":"Zhou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,6,18]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1603","DOI":"10.1212\/CON.0000000000000672","article-title":"Intracerebral hemorrhage","volume":"24","author":"Ziai","year":"2018","journal-title":"Continuum."},{"key":"B2","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1177\/17474930211065917","article-title":"World stroke organization (WSO): global stroke fact sheet 2022","volume":"17","author":"Feigin","year":"2022","journal-title":"Int J Stroke."},{"key":"B3","doi-asserted-by":"publisher","first-page":"S65","DOI":"10.1161\/STROKEAHA.113.000856","article-title":"Hemorrhagic stroke: introduction","volume":"44","author":"Hanley","year":"2013","journal-title":"Stroke."},{"key":"B4","doi-asserted-by":"publisher","first-page":"880","DOI":"10.1177\/23969873231191304","article-title":"Motor rehabilitation after stroke: European stroke organisation (ESO) consensus-based definition and guiding framework","volume":"8","author":"Kwakkel","year":"2023","journal-title":"Eur Stroke J."},{"key":"B5","doi-asserted-by":"publisher","first-page":"306","DOI":"10.1055\/s-0036-1581995","article-title":"Recovery and rehabilitation after intracerebral hemorrhage","volume":"36","author":"Saulle","year":"2016","journal-title":"Semin Neurol."},{"key":"B6","doi-asserted-by":"publisher","first-page":"3502","DOI":"10.1161\/STROKEAHA.114.005661","article-title":"Randomized controlled trial of early rehabilitation after intracerebral hemorrhage stroke: difference in outcomes within 6 months of stroke","volume":"45","author":"Liu","year":"2014","journal-title":"Stroke."},{"key":"B7","doi-asserted-by":"publisher","first-page":"1609","DOI":"10.3892\/etm.2013.1043","article-title":"Efficacy of hyperbaric oxygen treatment for depression in the convalescent stage following cerebral hemorrhage","volume":"5","author":"Cao","year":"2013","journal-title":"Exp Ther Med."},{"key":"B8","doi-asserted-by":"publisher","first-page":"862","DOI":"10.1007\/s12028-021-01311-3","article-title":"Neuroprotective therapies for spontaneous intracerebral hemorrhage","volume":"35","author":"Kearns","year":"2021","journal-title":"Neurocrit Care."},{"key":"B9","doi-asserted-by":"publisher","first-page":"223","DOI":"10.3390\/biomedicines11010223","article-title":"Acute spontaneous lobar cerebral hemorrhages present a different clinical profile and a more severe early prognosis than deep subcortical intracerebral hemorrhages-A hospital-based stroke registry study","volume":"11","author":"Mendiola","year":"2023","journal-title":"Biomedicines."},{"key":"B10","doi-asserted-by":"publisher","first-page":"125","DOI":"10.1177\/1747493018806196","article-title":"Frequency of post-stroke pneumonia: systematic review and meta-analysis of observational studies","volume":"14","author":"Badve","year":"2019","journal-title":"Int J Stroke."},{"key":"B11","doi-asserted-by":"publisher","first-page":"64","DOI":"10.3390\/ijms17010064","article-title":"Ischemia, immunosuppression and infection\u2013tackling the predicaments of post-stroke complications","volume":"17","author":"Shim","year":"2016","journal-title":"Int J Mol Sci."},{"key":"B12","doi-asserted-by":"publisher","first-page":"369","DOI":"10.1016\/j.juro.2015.10.039","article-title":"Re: the preventive antibiotics in stroke study (PASS): a pragmatic randomised open-label masked endpoint clinical trial","volume":"195","author":"Schaeffer","year":"2016","journal-title":"J Urol."},{"key":"B13","doi-asserted-by":"publisher","first-page":"116","DOI":"10.23937\/2378-3001\/1410116","article-title":"Clinical characteristics, predictors of outcome of patients admitted with intracerebral hemorrhage to benghazi medical center, Benghazi -Libya","volume":"9","author":"Sami","year":"2022","journal-title":"Int J Neurol Neurother."},{"key":"B14","doi-asserted-by":"publisher","first-page":"633","DOI":"10.1186\/s12877-023-04310-5","article-title":"Development and validation of a nomogram model for prediction of stroke-associated pneumonia associated with intracerebral hemorrhage","volume":"23","author":"Wang","year":"2023","journal-title":"BMC Geriatr."},{"key":"B15","doi-asserted-by":"publisher","first-page":"107444","DOI":"10.1016\/j.jstrokecerebrovasdis.2023.107444","article-title":"Development and validation of a nomogram for predicting pulmonary infections after intracerebral hemorrhage in elderly people","volume":"32","author":"Liu","year":"2023","journal-title":"J Stroke Cerebrovasc Dis."},{"key":"B16","doi-asserted-by":"publisher","first-page":"949","DOI":"10.1186\/s12879-020-05630-2","article-title":"The risk factors of intracranial infection in patients with intracerebral hemorrhage undergone hematoma puncture: what should we care","volume":"20","author":"Han","year":"2020","journal-title":"BMC Infect Dis."},{"key":"B17","doi-asserted-by":"publisher","first-page":"317","DOI":"10.2147\/CIA.S349978","article-title":"Application of machine learning algorithms to predict acute kidney injury in elderly orthopedic postoperative patients","volume":"17","author":"Chen","year":"2022","journal-title":"Clin Interv Aging."},{"key":"B18","doi-asserted-by":"publisher","first-page":"614204","DOI":"10.3389\/fcvm.2021.614204","article-title":"Development and validation of a predictive model for coronary artery disease using machine learning","volume":"8","author":"Wang","year":"2021","journal-title":"Front Cardiovasc Med."},{"key":"B19","doi-asserted-by":"publisher","first-page":"1976","DOI":"10.1002\/ecs2.1976","article-title":"Statistically reinforced machine learning for nonlinear patterns and variable interactions","volume":"8","author":"Ryo","year":"2017","journal-title":"Ecosphere."},{"key":"B20","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1186\/s40001-023-00995-x","article-title":"The prediction of in-hospital mortality in chronic kidney disease patients with coronary artery disease using machine learning models","volume":"28","author":"Ye","year":"2023","journal-title":"Eur J Med Res."},{"key":"B21","doi-asserted-by":"publisher","first-page":"e282","DOI":"10.1161\/STR.0000000000000407","article-title":"2022 guideline for the management of patients with spontaneous intracerebral hemorrhage: a guideline from the American heart association\/American stroke association","volume":"53","author":"Greenberg","year":"2022","journal-title":"Stroke."},{"key":"B22","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1097\/NRL.0000000000000269","article-title":"Stroke-associated pneumonia: a retrospective study of risk factors and outcomes","volume":"25","author":"Patel","year":"2020","journal-title":"Neurologist."},{"key":"B23","doi-asserted-by":"publisher","first-page":"e25213","DOI":"10.1097\/MD.0000000000025213","article-title":"Risk factors for and impact of poststroke pneumonia in patients with acute ischemic stroke","volume":"100","author":"Yuan","year":"2021","journal-title":"Medicine."},{"key":"B24","doi-asserted-by":"publisher","first-page":"1289","DOI":"10.1586\/14787210.2015.1085302","article-title":"Risk factors for lung infection in stroke patients: a meta-analysis of observational studies","volume":"13","author":"Yuan MZ Li","year":"2015","journal-title":"Expert Rev Anti Infect Ther."},{"key":"B25","doi-asserted-by":"publisher","first-page":"163","DOI":"10.1186\/s12933-024-02265-4","article-title":"Association between the stress hyperglycemia ratio and 28-day all-cause mortality in critically ill patients with sepsis: a retrospective cohort study and predictive model establishment based on machine learning","volume":"23","author":"Yan","year":"2024","journal-title":"Cardiovasc Diabetol."},{"key":"B26","doi-asserted-by":"publisher","first-page":"8761","DOI":"10.1021\/acs.jmedchem.9b01101","article-title":"Interpretation of compound activity predictions from complex machine learning models using local approximations and shapley values","volume":"63","author":"Rodriguez-Perez","year":"2020","journal-title":"J Med Chem."},{"key":"B27","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1097\/MCC.0000000000000535","article-title":"New guidelines for hospital-acquired pneumonia\/ventilator-associated pneumonia: USA vs. Europe","volume":"24","author":"Martin-Loeches","year":"2018","journal-title":"Curr Opin Crit Care."},{"key":"B28","doi-asserted-by":"publisher","first-page":"1533","DOI":"10.1164\/rccm.201001-0094OC","article-title":"Nosocomial pneumonia in the intensive care unit acquired by mechanically ventilated versus nonventilated patients","volume":"182","author":"Esperatti","year":"2010","journal-title":"Am J Respir Crit Care Med."},{"key":"B29","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1080\/01902148.2023.2264947","article-title":"Effect of mechanical ventilation under intubation on respiratory tract change of bacterial count and alteration of bacterial flora","volume":"49","author":"Xue-Meng","year":"2023","journal-title":"Exp Lung Res."},{"key":"B30","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1186\/s13054-021-03488-7","article-title":"Airway Pseudomonas aeruginosa density in mechanically ventilated patients: clinical impact and relation to therapeutic efficacy of antibiotics","volume":"25","author":"Migiyama","year":"2021","journal-title":"Crit Care."},{"key":"B31","doi-asserted-by":"publisher","first-page":"1337","DOI":"10.1378\/chest.126.4.1337","article-title":"The burden of methicillin resistant staphylococcus aureus in ventilator-associated pneumonia","volume":"126","author":"Shorr","year":"2004","journal-title":"Chest."},{"key":"B32","doi-asserted-by":"publisher","first-page":"495","DOI":"10.1007\/s12195-018-0537-8","article-title":"Inflammation and monocyte recruitment due to aging and mechanical stretch in alveolar epithelium are inhibited by the molecular chaperone 4-phenylbutyrate","volume":"11","author":"Valentine","year":"2018","journal-title":"Cell Mol Bioeng."},{"key":"B33","doi-asserted-by":"publisher","first-page":"1748","DOI":"10.3390\/nu14091748","article-title":"Nationwide prevalence and outcomes of long-term nasogastric tube placement in adults","volume":"14","author":"Hsu","year":"2022","journal-title":"Nutrients."},{"key":"B34","doi-asserted-by":"publisher","first-page":"932","DOI":"10.1001\/jamapediatrics.2014.696","article-title":"Changes in gastric and lung microflora with acid suppression: acid suppression and bacterial growth","volume":"168","author":"Rosen","year":"2014","journal-title":"JAMA Pediatr."},{"key":"B35","doi-asserted-by":"publisher","first-page":"395","DOI":"10.1038\/pr.2016.86","article-title":"Nasogastric feeding tubes from a neonatal department yield high concentrations of potentially pathogenic bacteria- even 1 d after insertion","volume":"80","author":"Petersen","year":"2016","journal-title":"Pediatr Res."},{"key":"B36","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1093\/gerona\/58.1.M52","article-title":"Pathogenic colonization of oral flora in frail elderly patients fed by nasogastric tube or percutaneous enterogastric tube","volume":"58","author":"Leibovitz","year":"2003","journal-title":"J Gerontol A Biol Sci Med Sci."},{"key":"B37","doi-asserted-by":"publisher","first-page":"508","DOI":"10.1179\/016164111X13007856084205","article-title":"Risk factors of stroke-associated pneumonia in Chinese patients","volume":"33","author":"Sui","year":"2013","journal-title":"Neurol Res."},{"key":"B38","doi-asserted-by":"publisher","first-page":"2964","DOI":"10.14260\/jemds\/2021\/606","article-title":"Predominance of bacterial flora causing lower respiratory tract infections following tracheostomy in patients of chettinad hospital & research institute, Chennai, India","volume":"10","author":"Thirumurthi","year":"2021","journal-title":"J Evol Med Dent Sci."},{"key":"B39","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1111\/ane.12095","article-title":"Predicting post-stroke pneumonia: the PANTHERIS score","volume":"128","author":"Harms","year":"2013","journal-title":"Acta Neurol Scand."},{"key":"B40","doi-asserted-by":"publisher","first-page":"416","DOI":"10.1007\/s12028-009-9325-6","article-title":"Usefulness of serum procalcitonin levels for the early diagnosis of stroke-associated respiratory tract infections","volume":"14","author":"Hug","year":"2011","journal-title":"Neurocrit Care."},{"key":"B41","first-page":"6163","article-title":"Serum iron and A(2)DS(2) score in stroke-associated pneumonia","volume":"2018","author":"Lu Y","year":"2015","journal-title":"Int J Clin Exp Med:."},{"key":"B42","doi-asserted-by":"publisher","first-page":"1155","DOI":"10.12669\/pjms.37.4.3932","article-title":"The value of serum procalcitonin in the anti-infection therapy of acute stroke patients","volume":"37","author":"Wang HL Li","year":"2021","journal-title":"Pak J Med Sci."},{"key":"B43","doi-asserted-by":"publisher","first-page":"CD008530","DOI":"10.1002\/14651858.CD008530.pub3","article-title":"Antibiotic therapy for preventing infections in people with acute stroke","volume":"1","author":"Vermeij","year":"2018","journal-title":"Cochrane Database Syst Rev."},{"key":"B44","doi-asserted-by":"publisher","first-page":"1835","DOI":"10.1016\/S0140-6736(15)00126-9","article-title":"Prophylactic antibiotics after acute stroke for reducing pneumonia in patients with dysphagia (STROKE-INF): a prospective, cluster-randomised, open-label, masked endpoint, controlled clinical trial","volume":"386","author":"Kalra","year":"2015","journal-title":"Lancet."},{"key":"B45","doi-asserted-by":"publisher","first-page":"e35023","DOI":"10.1097\/MD.0000000000035023","article-title":"Distribution of bacteria and risk factors in patients with multidrug-resistant pneumonia in a single center rehabilitation ward","volume":"102","author":"Wu","year":"2023","journal-title":"Medicine."},{"key":"B46","doi-asserted-by":"publisher","first-page":"581","DOI":"10.1007\/s12975-020-00863-4","article-title":"Disruptions of anaerobic gut bacteria are associated with stroke and post-stroke infection: a prospective case-control study","volume":"12","author":"Haak","year":"2021","journal-title":"Transl Stroke Res."},{"key":"B47","doi-asserted-by":"publisher","first-page":"1383","DOI":"10.1164\/rccm.201312-2291OC","article-title":"Propofol is associated with favorable outcomes compared with benzodiazepines in ventilated intensive care unit patients","volume":"189","author":"Lonardo","year":"2014","journal-title":"Am J Respir Crit Care Med."},{"key":"B48","doi-asserted-by":"publisher","first-page":"e10326","DOI":"10.7717\/peerj.10326","article-title":"Early abolition of cough reflex predicts mortality in deeply sedated brain-injured patients","volume":"8","author":"Kandelman","year":"2020","journal-title":"PeerJ."},{"key":"B49","doi-asserted-by":"publisher","first-page":"e14501","DOI":"10.1111\/nmo.14501","article-title":"Effects of dexmedetomidine on pharyngeal swallowing and esophageal motility-A double-blind randomized cross-over study in healthy volunteers","volume":"35","author":"Cajander","year":"2023","journal-title":"Neurogastroenterol Motil."},{"key":"B50","doi-asserted-by":"publisher","first-page":"758","DOI":"10.1016\/j.jcrc.2015.03.013","article-title":"Recovery of cough after extubation after coronary artery bypass grafting: a prospective study","volume":"30","author":"Kallesen","year":"2015","journal-title":"J Crit Care."},{"key":"B51","doi-asserted-by":"publisher","first-page":"CD004123","DOI":"10.1002\/14651858.CD004123","article-title":"Hyperbaric oxygen therapy for chronic wounds","volume":"2015","author":"Kranke","year":"2015","journal-title":"Cochrane Database Syst Rev."},{"key":"B52","doi-asserted-by":"publisher","first-page":"440","DOI":"10.1016\/j.biopha.2018.10.142","article-title":"Hyperbaric oxygen therapy: antimicrobial mechanisms and clinical application for infections","volume":"109","author":"Memar","year":"2019","journal-title":"Biomed Pharmacother."},{"key":"B53","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/j.freeradbiomed.2019.04.031","article-title":"Tissue-specific role of Nrf2 in the treatment of diabetic foot ulcers during hyperbaric oxygen therapy","volume":"138","author":"Dhamodharan","year":"2019","journal-title":"Free Radic Biol Med."},{"key":"B54","doi-asserted-by":"publisher","first-page":"1210","DOI":"10.3390\/biom11081210","article-title":"The effects of hyperbaric oxygenation on oxidative stress, inflammation and angiogenesis","volume":"11","author":"De Wolde","year":"2021","journal-title":"Biomolecules."},{"key":"B55","doi-asserted-by":"publisher","first-page":"958","DOI":"10.3390\/biom10060958","article-title":"The hyperoxic-hypoxic paradox","volume":"10","author":"Hadanny","year":"2020","journal-title":"Biomolecules."},{"key":"B56","volume-title":"Hyperbaric Oxygen 2003: Indications and Results: The Hyperbaric Oxygen Therapy Committee Report","year":"2003"},{"key":"B57","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1186\/1471-2377-7-32","article-title":"Thalamic haemorrhage vs internal capsule-basal ganglia haemorrhage: clinical profile and predictors of in-hospital mortality","volume":"7","author":"Arboix","year":"2007","journal-title":"BMC Neurol."}],"container-title":["Frontiers in Neurology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2025.1571755\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T05:46:47Z","timestamp":1750225607000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fneur.2025.1571755\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,18]]},"references-count":57,"alternative-id":["10.3389\/fneur.2025.1571755"],"URL":"https:\/\/doi.org\/10.3389\/fneur.2025.1571755","relation":{},"ISSN":["1664-2295"],"issn-type":[{"value":"1664-2295","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,18]]},"article-number":"1571755"}}