Exploiting Machine Learning Models for the Forewarning of Cardiovascular Risk in the Younger Demographic of India

Heart attacks together with cardiovascular diseases represent a serious global healthcare issue that particularly affects developing nations including India. Heart attack prevalence among youth has become alarmingly high while requiring strong predictive models for urgent medical intervention. Research compares the predictive effectiveness of machine learning approaches starting with Logistic Regression then continuing through Decision Trees to K-Nearest Neighbors (KNN) and Naive Bayes and Support Vector Machines (SVM) followed by Random Forest prior to Deep Neural Networks (DNN) in heart attack risk assessments of Indian young adults. A dataset including clinical and demographic variables which was used for model evaluation because data restrictions prevented adequate assessment of precision and F1-score and recall metrics. The comparison results between traditional machine learning methods and DNNs show valuable information about performance limitations that healthcare professionals together with decision-makers must consider for implementing data-driven heart disease prevention strategies.