Development and external validation of interpretable machine learning models for 7-, 14-, and 28-day mortality in sepsis-associated liver injury
- 1Critical Care Medicine Department, Xiaogan Hospital Affiliated to Wuhan University of Science and Technology , , ,
- 2Medical College, Wuhan University of Science and Technology , , ,
Abstract
Background: Mortality in sepsis-associated liver injury (SALI) remains high, and early risk stratification is difficult. Existing prognostic models commonly focus on a single endpoint, usually 28-day mortality, and rarely undergo independent external validation.
Methods: We developed machine-learning models using clinical features recorded during the first 24 h to predict all-cause in-hospital mortality within 7, 14, and 28 days. In MIMIC-IV (v3.1), 1,550 adults meeting the prespecified SALI criteria were allocated to a training cohort (n = 1,119) and a held-out internal validation cohort (n = 431). An independent Chinese cohort from Xiaogan Central Hospital (n = 229; 28-day in-hospital mortality, 45.9%) was used for external validation. One hundred candidate pipelines were evaluated using stratified 10-fold cross-validation. SHAP was used to describe feature attribution across prediction horizons.
Results: AUROCs in the held-out internal validation cohort were 0.840, 0.841, and 0.850 for 7-, 14-, and 28-day mortality, respectively; corresponding external AUROCs were 0.720, 0.717, and 0.763. The external 28-day model outperformed modified SOFA (0.763 vs. 0.673; P = 0.011) and MELD-Na (0.763 vs. 0.636; P = 0.002). A parsimonious logistic-regression sensitivity model achieved an external 28-day AUROC of 0.762, not significantly different from the full model (P = 0.970). The 14-day model was miscalibrated externally; recalibration improved the calibration slope from 0.563 to 0.970 but did not materially improve the Brier score. SHAP rankings varied across horizons and were interpreted descriptively rather than causally.
Conclusion: The models showed moderate transportability to one independent retrospective cohort. Multicentre prospective validation is needed before clinical implementation.
Related
- An interpretable machine learning model for predicting intensive care unit mortality risk in elderly patients with concomitant cardiogenic shock and sepsis: Development and external validation using two large critical care databases
- Development and validation of an interpretable machine learning model for early prediction of mortality in ICU patients with chronic obstructive pulmonary disease and heart failure
- Admission-time prediction of 7-day mortality in clinician-attributed hypertensive intracerebral hemorrhage: development and external validation against the ICH score
- Development and validation of an interpretable machine learning model for predicting in-hospital mortality among critically ill patients with liver cirrhosis
- High-volume cervical lymph node metastasis in papillary thyroid carcinoma: development and external validation of an interpretable prediction model.
BibTeX
@article{yao2026development,
title = {{Development and external validation of interpretable machine learning models for 7-, 14-, and 28-day mortality in sepsis-associated liver injury}},
author = {Hao Yao and Zhaopu Li and Jinxiao Lü and Wei Shen},
journal = {Frontiers in medicine},
year = {2026},
month = oct,
doi = {10.3389/fmed.2026.1942203},
url = {https://europepmc.org/article/PMC/PMC13719580},
note = {CC BY},
}
RIS
TY - JOUR
TI - Development and external validation of interpretable machine learning models for 7-, 14-, and 28-day mortality in sepsis-associated liver injury
AU - Hao Yao
AU - Zhaopu Li
AU - Jinxiao Lü
AU - Wei Shen
PY - 2026
DA - 2026/10/16
JO - Frontiers in medicine
DO - 10.3389/fmed.2026.1942203
UR - https://europepmc.org/article/PMC/PMC13719580
AB - Background: Mortality in sepsis-associated liver injury (SALI) remains high, and early risk stratification is difficult. Existing prognostic models commonly focus on a single endpoint, usually 28-day mortality, and rarely undergo independent external validation. Methods: We developed machine-learning models using clinical features recorded during the first 24 h to predict all-cause in-hospital mortality within 7, 14, and 28 days. In MIMIC-IV (v3.1), 1,550 adults meeting the prespecified SALI criteria were allocated to a training cohort (n = 1,119) and a held-out internal validation cohort (n = 431). An independent Chinese cohort from Xiaogan Central Hospital (n = 229; 28-day in-hospital mortality, 45.9%) was used for external validation. One hundred candidate pipelines were evaluated using stratified 10-fold cross-validation. SHAP was used to describe feature attribution across prediction horizons. Results: AUROCs in the held-out internal validation cohort were 0.840, 0.841, and 0.850 for 7-, 14-, and 28-day mortality, respectively; corresponding external AUROCs were 0.720, 0.717, and 0.763. The external 28-day model outperformed modified SOFA (0.763 vs. 0.673; P = 0.011) and MELD-Na (0.763 vs. 0.636; P = 0.002). A parsimonious logistic-regression sensitivity model achieved an external 28-day AUROC of 0.762, not significantly different from the full model (P = 0.970). The 14-day model was miscalibrated externally; recalibration improved the calibration slope from 0.563 to 0.970 but did not materially improve the Brier score. SHAP rankings varied across horizons and were interpreted descriptively rather than causally. Conclusion: The models showed moderate transportability to one independent retrospective cohort. Multicentre prospective validation is needed before clinical implementation.
SN - 2296-858X
ER -