HEART DISEASE PREDICTION MODEL - OPPOSITION BASED LEARNING AND WHALE OPTIMIZATION ALGORITHM WITH RNN CLASSIFIER

Abstract

Cardiovascular diseases (CVDs) remain a leading cause of mortality worldwide, emphasizing the importance of accurate prediction and early detection. This study aims to leverage cutting-edge machine learning algorithms to develop a robust predictive model for heart disease using a wide-ranging dataset obtained from the public UCI heart-disease dataset comprising 919 patients’ data and 14 attributes.  The dataset encompasses diverse patient characteristics, including demographic information, clinical attributes, and diagnostic tests, facilitating a holistic approach to prediction. Subsequently, comprehensive feature selection techniques were applied to refine the datasets, capturing nuanced relationships and enhancing the predictive capability. Feature Selection can improve the performance of machine learning models by creating relevant and informative features from raw data. By selecting features, Machine Learning models can make more accurate predictions, handle complex and distributed data, reduce overfitting, and extract valuable insights from categorical and numerical data. This proposed model focuses on the design of automated heart disease diagnosis model using Optimum Recurrent Neural Network (ORNN). The proposed model involves a feature selection approach using Whale Optimization algorithm and Opposition Based Learning scheme (OB-WOA) for identifying the optimal features. In addition, several classification models such as Support Vector Machine (SVM), Naïve Bayes (NB), k-Nearest Neighbours (k-NN), Decision Tree (DT), Extreme Gradient Boost (XG-Boost), and Recurrent Neural Networks (RNN). The Performance evaluation metrics including accuracy, precision, recall, and F1-score were employed to assess the models' predictive capabilities on UCI heart-disease dataset and the experimental results show that the proposed heart disease detection model OBL-WOA with RNN (ORNN) achieves the best accuracy for predicting the heart disease.