Accuracy is not enough: explainable boosting machine model and identification of candidate biomarkers for real-time sepsis risk assessment in the emergency department

Basit Öğe Kaydı
dc.contributor.authorYagin, Fatma Hilal
dc.contributor.authorAygun, Umran
dc.contributor.authorColak, Cemil
dc.contributor.authorAlkhalifa, Amal K.
dc.contributor.authorAlzakari, Sarah A.
dc.contributor.authorAghaei, Mohammadreza
dc.date.accessioned2026-04-04T13:33:13Z
dc.date.available2026-04-04T13:33:13Z
dc.date.issued2025
dc.departmentİnönü Üniversitesi
dc.description.abstractBackgroundSepsis poses a significant threat in emergency settings, necessitating tools for early and interpretable risk assessment. This study aimed to develop a robust explainable boosting machine (EBM) model, one of the explainable artificial intelligence (XAI) technologies, to construct a predictive model that balances high accuracy and clinical interpretability for use in emergency departments (EDs) and to examine candidate biomarkers.MethodsThe study identified a significant class imbalance problem in the sepsis distribution among 560 sepsis and 1012 non-sepsis patients. To address the imbalance issue, SMOTE-NC was applied in the training data. The data was divided into two parts, 80% training and 20% testing. To ensure the reliability of the models and to report unbiased results, this process was repeated 100 times and the average performance was reported. To determine the best model for sepsis prediction, five different models (AdaBoost, Gradient Boosting, CatBoost, LightGBM, and EBM) were trained, and their performances were evaluated. In the last stage, we presented local and global explanations of EBM.ResultsThe EBM model achieved the highest success by reaching 79.1% F1-score, 80.9% sensitivity, and 84.8% AUC after resampling. In the global explanations, the variables with the highest weights in the model's decision process were identified as positive blood culture, oxygen saturation, and procalcitonin, respectively.ConclusionThe EBM model accurately predicts sepsis risk based on clinically relevant biomarkers. The model's high performance and inherent transparency can foster trust among clinicians and facilitate its integration into emergency department workflows for real-time decision support.
dc.description.sponsorshipAlbert-Ludwigs-Universitt Freiburg im Breisgau (1016)
dc.description.sponsorshipOpen Access funding enabled and organized by Projekt DEAL.
dc.identifier.doi10.1186/s12873-025-01402-w
dc.identifier.issn1471-227X
dc.identifier.issue1
dc.identifier.pmid41315982
dc.identifier.scopus2-s2.0-105023334527
dc.identifier.scopusqualityN/A
dc.identifier.urihttps://doi.org/10.1186/s12873-025-01402-w
dc.identifier.urihttps://hdl.handle.net/11616/108985
dc.identifier.volume25
dc.identifier.wosWOS:001627997000003
dc.identifier.wosqualityQ1
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.indekslendigikaynakPubMed
dc.language.isoen
dc.publisherBmc
dc.relation.ispartofBmc Emergency Medicine
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/openAccess
dc.snmzKA_WOS_20250329
dc.subjectSepsis
dc.subjectMachine learning
dc.subjectExplainable artificial intelligence
dc.subjectExplainable boosting machine
dc.subjectBiomarker
dc.titleAccuracy is not enough: explainable boosting machine model and identification of candidate biomarkers for real-time sepsis risk assessment in the emergency department
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
PubMed İndeksli Yayın Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu