Identifying Heart Attack Risk in Vulnerable Population: A Machine Learning Approach

The COVID-19 pandemic has significantly increased the incidence of post-infection cardiovascular events, particularly myocardial infarction, in individuals over 40. While the underlying mechanisms remain elusive, this study employs a hybrid machine learning approach to analyze epidemiological data in assessing 13 key heart attack risk factors and their susceptibility. Based on a unique dataset that combines demographic, biochemical, ECG, and thallium stress tests, this study aims to design, develop, and deploy a clinical decision support system. Assimilating outcomes from five clustering techniques applied to the ‘Kaggle heart attack risk’ dataset, the study categorizes distinct subpopulations against varying risk profiles and then divides the population into ‘at-risk’ (AR) and ‘not-at-risk’ (NAR) groups using clustering algorithms. The GMM algorithm outperforms its competitors (with clustering accuracy and Silhouette coefficient scores of 84.24% and 0.2623, respectively). Subsequent analyses, employing Pearson correlation and linear regression as descriptors, reveal a strong association between the likelihood of experiencing a heart attack and the 13 risk factors studied, and these are statistically significant (p < 0.05). Our findings provide valuable insights into the development of targeted risk stratification and preventive strategies for high-risk individuals based on heart attack risk scores. The aggravated risk for postmenopausal patients indicates compromised individual risk factors due to estrogen depletion that may be further compromised by extraneous stress impacts, like anxiety and fear, aspects that have traditionally eluded data modeling predictions. The model can be repurposed to analyze the impact of COVID-19 on vulnerable populations.