Harris Hawks Optimization (HHO) Algorithm based on Artificial Neural Network for Heart Disease Diagnosis

Abstract

Signal processing methods usually diagnose heart disease, and the diagnosis of this type of disease by signal processing sometimes encounters many difficulties. To reduce diagnostic problems, careful feature selection and training are needed to analyze these signals. In this study, an attempt has been made to combine machine learning skills, such as neural network learning, with the Harris Hawks Optimization method to diagnose heart disease. In this paper, the heart disease diagnosis is analyzed with the feature selection method. For feature selection, the Harris Hawks Optimization Algorithm based on a fitting neural network is used. First, the Harris Hawks Optimization algorithm was implemented on the data, and the sample features were randomly selected. Then the sample features are trained by a neural network, and the best features are selected. Results show that the proposed method's accuracy, sensitivity, and precision for diagnosing heart disease are 92.75%, 92.15%, and 95.69%, respectively. The proposed method has a lower error in diagnosing heart disease from MLP, SVM, RF, and AdaBoost.