Deep Learning Networks Accurately Detect ST-Segment Elevation Myocardial Infarction and Culprit Vessel

Lin Wu; Guifang Huang; Xianguan Yu; Minzhong Ye; Lu Liu; Yesheng Ling; Xiangyu Liu; Dinghui Liu; Bin Zhou; Yong Liu; Jianrui Zheng; Suzhen Liang; Rui Pu; Xuemin He; Yanming Chen; Lanqing Han; Xiaoxian Qian

doi:10.3389/fcvm.2022.797207

Deep Learning Networks Accurately Detect ST-Segment Elevation Myocardial Infarction and Culprit Vessel

Front Cardiovasc Med. 2022 Mar 10:9:797207. doi: 10.3389/fcvm.2022.797207. eCollection 2022.

Authors

Lin Wu^{1

2}, Guifang Huang³, Xianguan Yu¹, Minzhong Ye⁴, Lu Liu⁵, Yesheng Ling¹, Xiangyu Liu⁶, Dinghui Liu¹, Bin Zhou¹, Yong Liu¹, Jianrui Zheng¹, Suzhen Liang¹, Rui Pu¹, Xuemin He², Yanming Chen², Lanqing Han³, Xiaoxian Qian¹

Affiliations

¹ Department of Cardiology, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China.
² Department of Endocrine and Metabolic Diseases, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China.
³ Center for Artificial Intelligence, Research Institute of Tsinghua, Pearl River Delta, Guangzhou, China.
⁴ Novelty-Checking Center, Guangdong Institute of Scientific and Technical Information, Guangzhou, China.
⁵ Department of Anesthesiology, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China.
⁶ School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China.

Abstract

Early diagnosis of acute ST-segment elevation myocardial infarction (STEMI) and early determination of the culprit vessel are associated with a better clinical outcome. We developed three deep learning (DL) models for detecting STEMIs and culprit vessels based on 12-lead electrocardiography (ECG) and compared them with conclusions of experienced doctors, including cardiologists, emergency physicians, and internists. After screening the coronary angiography (CAG) results, 883 cases (506 control and 377 STEMI) from internal and external datasets were enrolled for testing DL models. Convolutional neural network-long short-term memory (CNN-LSTM) (AUC: 0.99) performed better than CNN, LSTM, and doctors in detecting STEMI. Deep learning models (AUC: 0.96) performed similarly to experienced cardiologists and emergency physicians in discriminating the left anterior descending (LAD) artery. Regarding distinguishing RCA from LCX, DL models were comparable to doctors (AUC: 0.81). In summary, we developed ECG-based DL diagnosis systems to detect STEMI and predict culprit vessel occlusion, thus enhancing the accuracy and effectiveness of STEMI diagnosis.

Keywords: CNN-LSTM; ST-segment elevation myocardial infarction (STEMI); convolutional neural network (CNN); culprit vessel; deep learning (DL); electrocardiogram (ECG); long short-term memory (LSTM).