Heart Disease Feature Extraction Using Machine Learning

Abstract

In the realm of healthcare, accurate and efficient heart disease diagnosis and prediction are crucial tasks. This abstract introduces a study concentrating on feature extraction from heart disease data using an Enhanced Multi-Layer Perceptron (EMLP) neural network. The method combines deep learning with specialized enhancements such as dropout, batch normalization, and gradient clipping. These enhancements enhance accuracy and robustness in feature extraction. The trained enhanced MLP adeptly captures intricate patterns in heart disease data, generating informative and distinguishing features. These features contribute to precise disease diagnosis and prediction. Experimental results underscore the method's effectiveness, advancing heart disease detection and prognostication within medical data analysis.