What question did this study set out to answer?

To create an AI-based decision support system that predicts adverse outcomes in rhabdomyolysis patients.

February 19, 2026Informatics for Health and Social Care0 citations

Toward an explainable AI-Based clinical decision support system for predicting adverse outcomes in Rhabdomyolysis

FTFulden Cantaş TürkişDepartment of Biological Sciences BVBugra VarolIstanbul Aydın University YGYalcin GolcukMuğla University

Key Result

The CatBoost AI model predicted renal replacement therapy or 90-day mortality in rhabdomyolysis patients with AUC 0.942 and accuracy 91.3%, identifying key predictors.

Key Points

To create an AI-based decision support system that predicts adverse outcomes in rhabdomyolysis patients.
Developed a machine learning model using admission data from 1031 adults.
Applied multivariate imputation and Boruta feature selection for data preprocessing.
Utilized the CatBoost model for outcome prediction with high accuracy and discrimination.
Employed SHAP for identifying key predictors and validating clinical relevance.
Achieved an AUC of 0.942, indicating high discrimination ability of the model.
Demonstrated good accuracy at 0.913 and a Brier score of 0.080 for calibration.
Identified creatinine, troponin T, and albumin as significant predictors.
Confirmed superior net benefit of the model in decision-curve analysis.

Structured PICO

Does an explainable AI-based clinical decision support system accurately predict the composite outcome of renal replacement therapy or 90-day mortality in adults with rhabdomyolysis?

Population

1031 adults with rhabdomyolysis

Intervention

CatBoost machine learning model (explainable AI-based clinical decision support system) using routinely available admission data

Comparator

Treat-all or treat-none strategies (evaluated via decision-curve analysis)

Outcome

Composite outcome of renal replacement therapy or 90-day mortalitycomposite

An explainable AI model using routine admission data accurately predicts the need for renal replacement therapy or 90-day mortality in adults with rhabdomyolysis, providing a transparent tool for clinical decision support.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Rhabdomyolysis is a severe condition with high morbidity and mortality, driven by complications like acute kidney injury. Early risk stratification remains challenging as traditional scores fail to capture complex data patterns. This study lays the foundation for an explainable AI (XAI)-based clinical decision support system (CDSS) by developing a machine learning model to predict the composite outcome of renal replacement therapy or 90-day mortality. Using routinely available admission data from 1031 adults, we applied multivariate imputation, Boruta feature selection, and ADASYN for class imbalance. The CatBoost model achieved the highest discrimination (AUC = 0.942, 95% CI: 0.904-0.980), accuracy (0.913), and maintained good calibration (Brier score = 0.080). Shapley additive explanations (SHAP) identified creatinine, troponin T, and albumin as key predictors, validating clinical plausibility and enabling instance-level explanations for CDSS deployment. Decision-curve analysis confirmed superior net benefit against treat-all or treat-none strategies across clinically relevant thresholds. We propose a framework for integrating this interpretable model into electronic health records to provide real-time risk scores at the point of care. Calibration drift in older adults highlights the need for age-specific refinement, underscoring the value of transparent, evaluable AI in clinical informatics.

KI fragen

Bookmark

View Full Paper