Research Article ResNet-50 for 12-Lead Electrocardiogram Automated Diagnosis Nizar Sakli, 1,2 Haifa Ghabri, 2 Ben Othman Soufiene , 3 Faris. A. Almalki , 4 Hedi Sakli , 1,2 Obaid Ali , 5 and Mustapha Najjari 6 1 EITA Consulting, 5 Rue du Chant des Oiseaux, Montesson 78360, France 2 MACS Research Laboratory RL16ES22, National Engineering School of Gabes, Gabes University, Gabes 6029, Tunisia 3 PRINCE Laboratory Research, ISITcom, Hammam Sousse, University of Sousse, Sousse 4023, Tunisia 4 Department of Computer Engineering, College of Computers and Information Technology, Taif University, P.O. Box 11099, Taif 21944, Saudi Arabia 5 Ibb University, Department of Computer Science and Information Technology, Ibb, Yemen 6 LR18ES34 PEESE, National Engineering School of Gabes, Gabes University, Gabes 6029, Tunisia Correspondence should be addressed to Obaid Ali; obaid.alii2016@gmail.com Received 2 November 2021; Revised 13 January 2022; Accepted 22 March 2022; Published 28 April 2022 Academic Editor: Yassine Maleh Copyright © 2022 Nizar Sakli et al. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Nowadays, the implementation of Artificial Intelligence (AI) in medical diagnosis has attracted major attention within both the academic literature and industrial sector. AI would include deep learning (DL) models, where these models have been achieving a spectacular performance in healthcare applications. According to the World Health Organization (WHO), in 2020 there were around 25.6 million people who died from cardiovascular diseases (CVD). us, this paper aims to shad the light on cardiology since it is widely considered as one of the most important in medicine field. e paper develops an efficient DL model for automatic diagnosis of 12-lead electrocardiogram (ECG) signals with 27 classes, including 26 types of CVD and a normal sinus rhythm. e proposed model consists of Residual Neural Network (ResNet-50). An experimental work has been conducted using combined public databases from the USA, China, and Germany as a proof-of-concept. Simulation results of the proposed model have achieved an accuracy of 97.63% and a precision of 89.67%. e achieved results are validated against the actual values in the recent literature. 1. Introduction Nowadays, the medical field requires new techniques and technologies in order to evaluate information objectively. According to data from the World Health Organization (WHO), cardiovascular diseases (CVD) represent the leading cause of death globally, where the CVDs account for more than 30% of global mortality each year, and it is es- timated to reach around 130 million people by 2035 [1]. erefore, researchers are developing new methods for preventing, detecting, and treatment of diseases related to the CVD. ere are many types of cardiovascular abnor- malities, while this study focuses on 26 anomalies, which will be cited later. e electrocardiogram (ECG) is a recording of the electrical activity of the human heart, which is deemed as a noninvasiveness and real-time exam. It is still one of the essential pillars of the diagnosis of cardiac problems. In recent years, the methods of analysing CVDs have been strengthened by the introduction of imaging procedures, especially the echocardiogram. However, this does not change the importance and usefulness of ECGs, and the parameters could be extracted from this signal. e number of leads on a typical ECG acquisition equipment divides it into 1-lead, 3-lead, 6-lead, and 12-lead ECG. e 12-lead ECG is the most often utilized kind in clinical practice due to its ability to concurrently capture the potential changes of 12 sets of electrode patches attached to the body in standardized places [2]. When comparing to other types of ECG acqui- sition equipment, 12-lead ECG provides more information on cardiac activity and is frequently utilized in hospital for diagnosis and treatment. In fact, many essential parameters Hindawi Computational Intelligence and Neuroscience Volume 2022, Article ID 7617551, 16 pages https://doi.org/10.1155/2022/7617551