An In-Depth Analysis of Machine Learning Algorithms for Disease Detection and Classification

Abstract

Heart Disease (HD) is a widespread health concern affecting millions of individuals worldwide, with common symptoms including breathlessness, swollen feet, and muscle weakness. However, current diagnostic methods for heart disease often need to catch up in early detection, primarily due to accuracy and processing time issues. As a result, researchers are continuously striving to develop more effective solutions for identifying heart conditions at an earlier stage. This study introduces a machine learning-based approach aimed at providing rapid and accurate heart disease diagnosis. The system utilizes a combination of classification techniques, such as Support Vector Machines (SVM), Decision Trees, and Random Forests. A hybrid algorithm that integrates these techniques is developed to enhance the model's accuracy further. The performance of the classifiers is assessed using various evaluation metrics to measure their effectiveness.

Additionally, feature selection methods are employed to refine the models, ensuring that only the most relevant attributes are considered. The final diagnosis is delivered through a user-friendly Flask framework-based website. Furthermore, the proposed method can generate personalized food recommendations for patients diagnosed with heart disease, contributing to their overall health management and well-being.