Prediction of In-Hospital Mortality in Non-ST-Segment Elevation Myocardial Infarction, Based on Interpretable Machine Learning

• Published in: Applied sciences Vol. 15; no. 8; p. 4226
• Main Authors: Cao, Mengru, Li, Chunhui
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.04.2025
• Subjects: Accuracy; Acute coronary syndromes; Algorithms; Analysis; Cholesterol; Data mining; Decision making; Decision trees; Disease; Enzymes; Heart attack; Heart attacks; Hemoglobin; Hospitalization; Hospitals; in-hospital mortality; Kinases; Length of stay; Machine learning; Medical advice systems; Missing data; Mortality; non-ST-elevation myocardial infarction; Patient outcomes; Patients; prediction model; Sampling techniques; Shapley additive explanations; Structured Query Language-SQL; Support vector machines; Taiwan; Variables; Taiwan
• DOI: 10.3390/app15084226
• ISSN: 2076-3417

This study sought to establish machine learning models for forecasting in-hospital mortality in non-ST-segment elevation myocardial infarction (NSTEMI) patients, and focused on model interpretability using Shapley additive explanations (SHAP). Data were gathered from the Medical Information Mart for Intensive Care—IV database. The synthetic minority over-sampling technique and Edited Nearest Neighbors were used to address class imbalance. Four machine learning algorithms were employed, including Adaptive Boosting (AdaBoost), Random Forest (RF), Gradient Boosting Decision Trees (GBDT), and eXtreme Gradient Boosting (XGBoost). SHAP was utilized to improve transparency and credibility. The all-features RF model demonstrated optimal performance, with an accuracy of 0.8513, precision of 0.9016, and AUC of 0.8903. The SHAP summary plot for the RF model revealed that Acute Physiology Score III, lactate dehydrogenase, and lactate were the three most crucial characteristics, with higher values indicating a greater risk. The study demonstrates the applicability of machine learning, particularly RF, in predicting in-hospital mortality for NSTEMI patients, with the use of SHAP enhancing model interpretability and providing clinicians with clearer insights into feature contributions.