Abstract / Summary
Abstract Background Acute coronary syndrome (ACS), including ST-elevation myocardial infarction (STEMI) and non-ST-elevation myocardial infarction/unstable angina pectoris (NSTEMI/USAP) is a major cause of cardiovascular morbidity and mortality. Objective To determine clinical and biochemical predictors of ACS type and how well they predict using statistical and machine learning methods. Methodology A Case-Control study was performed at the Hayatabad Medical Complex, Peshawar, between November 2024 and March 2025 with 100 patients participating in the study with an equal number of STEMI and NSTEMI/USAP patients. IBM SPSS software version 27 was used to analyze the data. The Shapiro-Wilk test was used to determine the normality of the data; the Mann-Whitney U test and Chi-square test were used to compare the groups. Binary logistic regression was used to screen the predictors. Supervised machine learning models ( random forest, support vector machine, k nearest neighbors) were trained to classify ACS type. Results There were statistically significant group differences in oxygen saturation and HDL. The main predictors were vitamin D levels, HDL, oxygen saturation and dyslipidemia. Machine learning models exhibited moderate performance with Logistic Regression and Support Vector Machine having the highest accuracy and Random Forest having an AUC of 0.792. Furthermore, vitamin D deficiency was observed more frequently among STEMI patients compared to NSTEMI/USAP patients. Conclusion Machine learning models moderately predicted ACS classification, with vitamin D deficiency, dyslipidemia, and oxygen saturation identified as key predictors. Vitamin D deficiency was more common in STEMI patients, suggesting a role in ACS severity. Trial registration not applicable