Interpretable gradient boosting machine model for predicting in-hospital mortality in sepsis-induced myocardial injury: a multicenter development, validation, and web-based clinical implementation - Takeaways - MDSpire

Interpretable gradient boosting machine model for predicting in-hospital mortality in sepsis-induced myocardial injury: a multicenter development, validation, and web-based clinical implementation

  • By

  • Lina Chen

  • Qianru Yuan

  • Yitong Ma

  • July 1, 2026

  • 0 min

Share

  • 1

    Sepsis-induced myocardial injury (SIMI) is a severe complication of sepsis, with in-hospital mortality rates reaching 35%.

  • 2

    The Gradient Boosting Machine (GBM) model achieved an internal validation AUC of 0.751, outperforming other evaluated algorithms.

  • 3

    External validation yielded AUCs of 0.924 in eICU and 0.703 in a Chinese cohort, confirming the model's generalizability.

  • 4

    Key predictors for in-hospital mortality identified by SHAP included APS III Score, Hypertension, and SOFA Score.

  • 5

    A simplified model with five variables was deployed on a user-friendly platform for real-time risk assessment in clinical settings.

Original Source(s)

Interpretable gradient boosting machine model for predicting in-hospital mortality in sepsis-induced myocardial injury: a multicenter development, validation, and web-based clinical implementation

  • by Lina Chen, Qianru Yuan, Yitong Ma

  • July 1, 2026

Related Content

Beyond cardiac fibroblasts: research advances on understanding and targeting intercellular communication networks in cardiac fibrosis

Progression of mitral and tricuspid regurgitation in patients with and without atrial fibrillation

The characteristics and clinical outcomes of atrial fibrillation patient in middle income country of Indonesia