Predictive modeling for heart attack detection in Rwanda: A comprehensive analysis using R programming machine learning models

Heart disease remains a major health concern worldwide, leading to significant morbidity and mortality. It has become increasingly important to develop accurate prediction models for early detection and prevention of heart attacks. Several studies have demonstrated the potential of data science and machine learning algorithms in predicting heart disease. Rwanda faces a growing burden of cardiovascular disease, with heart attacks posing a significant public health threat. This study aims to develop and evaluate predictive models for heart attack detection in Rwanda using R programming and probability assessment techniques. Datasets capturing clinical, demographic, and lifestyle factors such as age, sex, blood pressure, cholesterol, and blood sugar, our model aims to achieve an accuracy of 70% in predicting heart attacks are analyzed to identify individual risk profiles and improve early diagnosis. This comprehensive analysis explores various R- based algorithms such as linear regression, focusing on their accuracy, sensitivity, specificity, and clinical interpretability. Additionally, we employ probability assessment methods to quantify individual heart attack risk and inform personalized preventive strategies. Transparent decision-making processes and visualizations enhance the interpretability of the model, facilitating its potential integration into Rwandan clinical settings. The predictive model will correctly identify 97% of people who have had a heart attack (TPR = 0.97), but it will also incorrectly identify 3% of people who have not had a heart attack (FPR = 0.03).