Comprehensive electrocardiographic diagnosis based on deep learning

• Published in: Artificial intelligence in medicine Vol. 103; p. 101789
• Main Authors: Lih, Oh Shu, Jahmunah, V, San, Tan Ru, Ciaccio, Edward J, Yamakawa, Toshitaka, Tanabe, Masayuki, Kobayashi, Makiko, Faust, Oliver, Acharya, U Rajendra
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.03.2020
• Subjects: 10-fold validation; Cardiovascular diseases; Congestive heart failure; Convolutional neural network; Coronary artery disease; Coronary Artery Disease - diagnostic imaging; Coronary Artery Disease - pathology; Deep Learning; Electrocardiography - methods; Heart Diseases - diagnosis; Heart Diseases - diagnostic imaging; Heart Diseases - pathology; Heart Failure - diagnostic imaging; Heart Failure - pathology; Humans; Long short-term memory; Myocardial infarction; Myocardial Infarction - diagnostic imaging; Myocardial Infarction - pathology; Neural Networks, Computer; Signal Processing, Computer-Assisted; Myocardial infarction; Deep learning; 10-fold validation; Convolutional neural network; Long short-term memory; Congestive heart failure; Cardiovascular diseases; Coronary artery disease
• ISSN: 0933-3657; 1873-2860; 1873-2860
• DOI: 10.1016/j.artmed.2019.101789

•Deep learning techniques for classification of MI, CAD and CHF are discussed.•First study to present deep learning technique for 4-class classification.•CNN coupled with LSTM yielded a high accuracy of 98.51%.•Future work using deep learning to detect early stages of CAD is proposed. Cardiovascular disease (CVD) is the leading cause of death worldwide, and coronary artery disease (CAD) is a major contributor. Early-stage CAD can progress if undiagnosed and left untreated, leading to myocardial infarction (MI) that may induce irreversible heart muscle damage, resulting in heart chamber remodeling and eventual congestive heart failure (CHF). Electrocardiography (ECG) signals can be useful to detect established MI, and may also be helpful for early diagnosis of CAD. For the latter especially, the ECG perturbations can be subtle and potentially misclassified during manual interpretation and/or when analyzed by traditional algorithms found in ECG instrumentation. For automated diagnostic systems (ADS), deep learning techniques are favored over conventional machine learning techniques, due to the automatic feature extraction and selection processes involved. This paper highlights various deep learning algorithms exploited for the classification of ECG signals into CAD, MI, and CHF conditions. The Convolutional Neural Network (CNN), followed by combined CNN and Long Short-Term Memory (LSTM) models, appear to be the most useful architectures for classification. A 16-layer LSTM model was developed in our study and validated using 10-fold cross-validation. A classification accuracy of 98.5% was achieved. Our proposed model has the potential to be a useful diagnostic tool in hospitals for the classification of abnormal ECG signals.