Novel Implementation of Heart Disease Classification Model using RNN Classification

Heart disease is the major health issue or challenge faced by the entire world in modern medicine. It has become a crucial factor for increasing the mortality rate. The desperation of heart disease is more vital and can even result in vulnerable consequences if it is not predicted at the initial stage. The methods such as electronic health records, monitoring body are network continuously and diagnosing the patient health condition via the medical sensors projection and wearable device across human bodies. Since the generated data from the human body are continuous and huge in amount, the data mining techniques are utilised for efficient classification of obtained health data. Moreover, the classification of health data is the most critical process as it needs an accurate execution with the early detection of heart disease. As most medical enthusiasts and practicing physicians finding reveals that difficult to diagnose the disease at an early stage is the key for the failures of incurable disease. Hence it is important to diagnose patients early to save life is the most challenging task for the medical fraternity. This research proposes to reduce the risk of heart diseases by effective feature selection and classification based prediction system to predict heart diseases. Therefore, the earlier prediction of heart disease with higher accuracy is an important limitation behind every existing process. So, this research attempts to develop an efficient classifier and predicts heart disease at the initial stage with highperformance measures and accuracy. The significant contribution of this research is divided into three parts. First, an effective method is implemented to predict heart disease by feature selection and classification. The proposed research comprises of optimised unsupervised technique for feature selection and novel MLP-EBMDA (Multi-Layer Perceptron for Enhanced Brownian Motion-based Dragonfly Algorithm) for classification for heart disease prediction. In this implementation, the input will be obtained from the dataset and performed pre-processing followed by the proposed feature selection technique that efficiently performs the selection of features. Based on selected features, heart disease classification vi is performed using novel MLP-EBMDA that helps to predict heart disease at the initial stage. This suggested system possesses an accuracy rate of 94.28 percentage in predicting heart disease as normal or abnormal in an effective way.