A Prognostic Prediction Model for Elderly Sepsis Based on Transformer and Explainable Artificial Intelligence: A Multicenter Study

Corresponding author: XIE Jian, xiejian0518@sina.com

DOI: 10.12201/bmr.202605.00087

•  

  Abstract: Purpose/Significance The prognosis of elderly sepsis is highly heterogeneous, and existing prediction tools lack both accuracy and interpretability. This study aims to develop and validate an interpretable prognostic prediction model specifically for elderly patients with sepsis. Methods/Process Elderly patients (≥60 years) meeting the Sepsis?3.0 criteria were retrospectively enrolled from the eICU?CRD (training set) and MIMIC?IV (external validation set) databases. Thirteen core inflammatory indicators were extracted, and an explainable sepsis prognosis transformer (ESPT) model was constructed based on the Transformer architecture. The SHAP framework was employed for global feature importance ranking and visual interpretation of individual predictions. Results/Conclusion The ESPT model achieved an area under the curve (AUC) of 0.8589 in the training set and 0.8147 in the external validation set, outperforming traditional machine learning models such as Random Forest and XGBoost. SHAP analysis identified blood glucose, hemoglobin, and hematocrit as key prognostic drivers. The ESPT model can serve as an effective auxiliary tool for risk stratification and clinical decision making in elderly patients with sepsis.

  Key words: Sepsis; Prognostic model; Explainable artificial intelligence; Inflammatory biomarkers; Multicenter study

  Submit time: 25 May 2026

  Copyright: The copyright holder for this preprint is the author/funder, who has granted biomedRxiv a license to display the preprint in perpetuity.
• html

• 图表

• Current Research Status on Magnesium Ion Levels and Prognosis of Sepsis. 2025. doi: 10.12201/bmr.202509.00008

  Qin Chao. Progress in the study of the effect of intestinal microecological changes on the prognosis of elderly sepsis patients. 2025. doi: 10.12201/bmr.202501.00041

  XIANG AININGKUN, TIAN JINGXUE, HU DEHUA, LIU HAIXIA. Comparative Study on Response Efficacy of Generative Artificial Intelligence for Elderly Diabetes Mellitus. 2025. doi: 10.12201/bmr.202503.00019

  GE Xiaoling. Application of Artificial Intelligence Large Models in Healthcare:a Survey. 2024. doi: 10.12201/bmr.202408.00039

  钱新阳, wangyumin. Prognostic Value of pSOFA Score Combined with AGI Grade in Children with Sepsis. 2024. doi: 10.12201/bmr.202411.00002

  zhoupengyang, liulili, chensixuan, zhangbaolin, puxuhong. Construction and Practice of Connotation Quality Control System of Traditional Chinese Medicine Electronic Medical Record Based on Artificial Intelligence Large Model. 2026. doi: 10.12201/bmr.202605.00088

  XING Qingmin, XIAO Tao, LI Ming. Prognostic Value of the Lactate/Albumin Ratio Combined with the APACHE II Score in Patients with Sepsis. 2026. doi: 10.12201/bmr.202603.00104

  Ma Xiang. Heart Sound and Cardiovascular Diseases under the Background of Artificial Intelligence + Medical and Health Care. 2026. doi: 10.12201/bmr.202601.00055

  ZhengYanli, Han Fuhai, LI Shuyu, SU Wenxing. Application Status and Prospect of Artificial Intelligence Large Models in Medicine. 2023. doi: 10.12201/bmr.202312.00027

  Huanghanqi, Zhoumanci, Yanghuiwen, Wangruizhi, Wangruojia. A Study of Ethical Risks of Medical Artificial Intelligence Applications from the Perspective of the Public in Society. 2025. doi: 10.12201/bmr.202511.00062

• ID	Submit time	Number	Download
  1	2025-11-29	10.12201/bmr.202605.00087V1	Download

• Public  Anonymous  To author only Submit Comments