AI Techniques for Efficient Healthcare Systems in ECG Wave Based Cardiac Disease Detection by High Performance Modelling

Abstract

Heart disease (HD) is extremely lethal by nature and claims a disproportionately large number of lives worldwide. Early and reliable detection techniques are necessary to prevent fatalities from HD. Clinical test results, electrocardiogram (ECG) signal, the heart sound signal, impedance cardiography (ICG), magnetic resonance imaging, and computer tomography (CT) can all be used to determine whether an individual has HD. This research propose novel technique in efficient healthcare system by ECG wave based cardiac disease detection using deep learning architecture with high performance modelling. Here the input is collected as ECG waves which has been processed and obtained as ECG wave fragments. This ECG fragment features has been extracted using deep belief kernel principal neural network. Based on this extracted features the patients 3D heart image has been collected and classified using deep quantum multilayer convolutional neural networks. Here the experimental analysis has been carried out in terms of accuracy, precision, recall, F-score, SNR, RMSE. Proposed technique attained accuracy of 95%, precision of 81%, recall of 69%, F-1score of 73%, SNR of 59% and RMSE of 62%.