Machine learning model for ACLF risk in decompensated cirrhosis.
- 1Department of Infectious Diseases, Henan Provincial People's Hospital, Zhengzhou University People's Hospital, Zhengzhou, Henan, China.
Abstract
Objective
To develop a machine learning-based predictive model for acute-on-chronic liver failure (ACLF) risk in patients with decompensated cirrhosis using baseline clinical indicators at hospital admission.Materials and methods
This retrospective cohort included 360 decompensated cirrhosis patients with intensive monitoring needs or infection/circulatory risk factors, randomly divided into training and validation sets at a 7:3 ratio. Three models-random forest, support vector machine, and logistic regression-were constructed. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) were employed for model interpretability.Results
In the training set, 52 patients (20.6%) experienced ACLF. Multivariate analysis identified total bilirubin (TBil) (OR = 1.051, 95% CI: 1.030-1.073, P < 0.001), arterial lactate (OR = 1.341, 95% CI: 1.058-1.700, P = 0.015), and white blood cell count (WBC) (OR = 1.111, 95% CI: 1.007-1.227, P = 0.036) as independent risk factors, whereas prothrombin time activity (PTA) (OR = 0.961, 95% CI: 0.941-0.982, P < 0.001) was protective. The random forest model showed comparable performance to logistic regression with training/validation AUCs of 0.833 (95% CI: 0.757-0.910) and 0.700 (95% CI: 0.555-0.845). All models showed acceptable discrimination and net clinical benefit on DCA. Calibration was good in the training set but moderate in the validation set (calibration slopes 0.580-0.620). SHAP importance ranking placed PTA, TBil, WBC, and lactate as the top predictors.Conclusion
The random forest model showed acceptable training performance, but the wide validation AUC confidence interval suggests limited stability, likely due to small sample size and few events. External validation is needed before clinical use. Nonetheless, the model shows potential for early warning and risk stratification in decompensated cirrhosis.
Related
- Dual-pathway targeted therapy for decompensated cirrhosis: Combined transjugular intrahepatic portosystemic shunt and cyclooxygenase-2 inhibition
- Acute Kidney Disease and Clinical Outcomes in Stable Outpatients With Decompensated Cirrhosis: A Prospective Cohort Study
- Clinical and patient-reported predictors of days alive and out of hospital in decompensated cirrhosis
- Plasma Cystatin C as a Marker of Acute-on-Chronic Liver Failure and 28-Day Mortality in Patients With Acutely Decompensated Liver Cirrhosis: A Prospective Observational Study
- Randomised, double-blind, placebo-controlled trial of faecal microbiota transplantation in patients with decompensated cirrhosis: the Livergut protocol.
BibTeX
@article{chen2026machine,
title = {{Machine learning model for ACLF risk in decompensated cirrhosis.}},
author = {Liwen Chen and Peipei Ren and Yi Kang},
journal = {Frontiers in medicine},
year = {2026},
month = oct,
doi = {10.3389/fmed.2026.1948234},
url = {https://europepmc.org/article/MED/42908680},
note = {CC BY},
}
RIS
TY - JOUR
TI - Machine learning model for ACLF risk in decompensated cirrhosis.
AU - Liwen Chen
AU - Peipei Ren
AU - Yi Kang
PY - 2026
DA - 2026/10/07
JO - Frontiers in medicine
DO - 10.3389/fmed.2026.1948234
UR - https://europepmc.org/article/MED/42908680
AB - Objective To develop a machine learning-based predictive model for acute-on-chronic liver failure (ACLF) risk in patients with decompensated cirrhosis using baseline clinical indicators at hospital admission.Materials and methods This retrospective cohort included 360 decompensated cirrhosis patients with intensive monitoring needs or infection/circulatory risk factors, randomly divided into training and validation sets at a 7:3 ratio. Three models-random forest, support vector machine, and logistic regression-were constructed. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) were employed for model interpretability.Results In the training set, 52 patients (20.6%) experienced ACLF. Multivariate analysis identified total bilirubin (TBil) (OR = 1.051, 95% CI: 1.030-1.073, P < 0.001), arterial lactate (OR = 1.341, 95% CI: 1.058-1.700, P = 0.015), and white blood cell count (WBC) (OR = 1.111, 95% CI: 1.007-1.227, P = 0.036) as independent risk factors, whereas prothrombin time activity (PTA) (OR = 0.961, 95% CI: 0.941-0.982, P < 0.001) was protective. The random forest model showed comparable performance to logistic regression with training/validation AUCs of 0.833 (95% CI: 0.757-0.910) and 0.700 (95% CI: 0.555-0.845). All models showed acceptable discrimination and net clinical benefit on DCA. Calibration was good in the training set but moderate in the validation set (calibration slopes 0.580-0.620). SHAP importance ranking placed PTA, TBil, WBC, and lactate as the top predictors.Conclusion The random forest model showed acceptable training performance, but the wide validation AUC confidence interval suggests limited stability, likely due to small sample size and few events. External validation is needed before clinical use. Nonetheless, the model shows potential for early warning and risk stratification in decompensated cirrhosis.
SN - 2296-858X
ER -