Heart Disease Prediction using Machine Learning and Deep Learning

Heart disease is a leading cause of mortality globally, pointing towards the necessity of effective screening and predictive functions to support early detection and treatment. Based on this research study, an emphasis is placed on developing and executing a robust predictive model for heart disease through the integration of machine learning (ML) and deep learning (DL) methods. Particularly, we utilize Logistic Regression, Support Vector Machine and Random Forest as ML classifiers, while Convolution Neural Networks and Artificial Neural Networks are utilized as DL models for clinical data set analysis and prediction of heart disease risk. For improved performance, we combine state-of-the-art feature extraction techniques with these models to enhance predictive accuracy and model inter predictability. Our experimental results identify that highest performance was achieved by the Random Forest classifier at accuracy of 92% and next by the CNN model with accuracy of 91%, highlighting the strength of deep learning and ensemble methods to pull out subtle patterns from data. Through the use of such algorithms, our research adds to the body of literature in favor of AI-driven solutions in medical diagnosis to improve patient outcomes. Key Words: Heart Disease Prediction, Machine Learning, Deep Learning, Classification Models, Neural Network.