Predicting mortality amongst Jordanian men with heart attacks using the chi-square automatic interaction detection...

Area of Science:

• Cardiovascular Disease Research
• Machine Learning in Healthcare
• Public Health Informatics

Background:

• Heart attack is a leading global cardiovascular disease.
• Men are disproportionately affected by cardiac conditions.
• Accurate prediction of heart attack outcomes is crucial for saving lives.

Purpose of the Study:

• To propose and evaluate the Chi-Squared Automated Interactive Detection (CHAID) model for predicting mortality in men experiencing heart attacks.
• To identify significant risk factors for heart attack mortality in a Jordanian male population.

Main Methods:

• Retrospective, predictive study design using data from Jordan's electronic health solution system (2015-2021).
• Information gathered from male patients admitted to public hospitals.
• Application of the CHAID algorithm for predictive modeling.

Main Results:

• The CHAID model demonstrated high predictive accuracy (93.72%) and an Area Under the Curve (AUC) of 0.792.
• Key predictors of mortality included governorates, age, pulse oximetry, medical diagnosis, pulse pressure, heart rate, and systolic blood pressure.
• The CHAID model was identified as the top-performing model for predicting mortality in this cohort.

Conclusions:

• Demographic characteristics and hemodynamic readings are significant predictors of heart attack mortality.
• Machine learning algorithms, specifically the CHAID model, can effectively predict mortality risk in heart attack patients.
• The findings highlight the utility of the CHAID model for improving patient outcomes in cardiovascular disease management.