Interpretable Machine Learning Model for Early Mortality Prediction in Septic Patients Using Routine Post-Diagnosis Clinical Data: A Multicenter Study
We developed a machine learning model to predict the risk of early 7-day mortality in sepsis patients based on routine clinical data obtained immediately after diagnosis and validated its potential as a clinically reliable tool, achieving an AUROC of 0.767 in the training set. The use of SHAP-based interpretation enhances model interpretability, enabling clinicians to better understand the factors influencing mortality, identify high-risk patients early, and implement timely interventions to improve outcomes.
Authors
- Zaiqian Che (ORCID: https://orcid.org/0009-0007-6567-825X)
- Wenwu Sun (ORCID: https://orcid.org/0000-0001-5866-5150)
- Shu Wang (ORCID: https://orcid.org/0000-0002-7099-2514)
- Yang Li (ORCID: https://orcid.org/0000-0001-5848-3461)
- Zhang Li-juan (ORCID: https://orcid.org/0000-0002-5164-6819)
- Bing Zhao (ORCID: https://orcid.org/0000-0002-6955-8446)
- Dan Mou
Institutions
- Army Medical University (CN)
- Shanghai Jiao Tong University (CN)
- Ruijin Hospital (CN)
- Daping Hospital (CN)
- Chongqing Emergency Medical Center (CN)
- Huazhong University of Science and Technology (CN)
Publication Details
- Journal
- Journal of Inflammation Research
- Published
- 2025-10-01
- DOI
- https://doi.org/10.2147/jir.s553042
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Chongqing University
- Huazhong University of Science and Technology
- Tongji University
- Shanghai Jiao Tong University
- Army Medical University
- Ruijin Hospital
- School of Medicine, Shanghai Jiao Tong University
- Tongji Medical College, Huazhong University of Science and Technology