Automatic heart disease class detection using convolutional neural network architecture‐based various optimizers‐networks. Issue 1 (30th March 2021)

• Record Type: Journal Article
• Authors: Fradi, Marwa; Khriji, Lazhar; Machhout, Mohsen; Hossen, Abdulnasir
• Is Part Of: IET smart cities. Volume 3:Issue 1(2021)
• Page Start: 3
• Page End: 15
• Publication Date: 2021-03-30
• Subjects: Smart cities -- Periodicals; City planning -- Technological innovations -- Periodicals; Cities and towns -- Growth -- Periodicals; Sustainable urban development -- Periodicals; Sustainable development; City planning -- Technological innovations; Cities and towns -- Growth; Periodicals; 307.76
• DOI: 10.1049/smc2.12003
• Languages: English
• ISSNs: 2631-7680

Abstract

Abstract: Early heart disease class detection is of great interest to reduce the mortality rate. In this context, computational techniques have been proposed to solve this issue. Thus, here, a deep learning architecture is proposed to automatically classify the patient's Electrocardiogram (ECG) signal into a specific class according to the ANSI–AAMI standards. The proposed methodology is a multi‐stage technique. The first stage combines an R–R peak extraction with a low‐pass filter applied on the ECG data for noise removal. The second stage shows the proposed convolutional neural network‐based fully connected layers architecture, using different network optimizers. Different ECG databases, including challenging tasks, have been used for validation purpose. The whole system is implemented on both CPU and GPU for complexity analysis. For the predicted improved PTB data set, the classification accuracy results achieve 99.37%, 99.15% and 99.31% for training, validation and testing, respectively. Besides, for the MIT‐BIH‐database, the training, validation and testing accuracies are 99.5%, 99.06% and 99.34%, respectively. A top F1‐score of 0.99 is obtained. Experimental results show a high achievement compared to the state‐of‐the‐art models where the implementation of GPU confirms the low computational complexity of the system.