ML–ResNet: A novel network to detect and locate myocardial infarction using 12 leads ECG. (March 2020)

• Record Type: Journal Article
• Authors: Han, Chuang; Shi, Li
• Is Part Of: Computer methods and programs in biomedicine. Volume 185(2020)
• Publication Date: 2020-03
• Subjects: Myocardial infarction -- Convolutional neural network -- Residual blocks -- Feature fusion -- Automated detection -- Inter-patient scheme; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2019.105138
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: A novel ML-ResNet model including three residual blocks and multi-lead feature fusion is employed to detect and locate MI. The single lead feature branch network can learn representative features of different levels between different layers. The model displays more vital clinical significance under inter-patient schemes. Compared with published literatures, the proposed model for MI detection and location has achieved the best results. Abstract: Background and objective: Myocardial infarction (MI) is one of the most threatening cardiovascular diseases for human beings, which can be diagnosed by electrocardiogram (ECG). Automated detection methods based on ECG focus on extracting handcrafted features. However, limited by the performance of traditional methods and individual differences between patients, it's difficult for predesigned features to detect MI with high performance. Methods: The paper presents a novel method to detect and locate MI combining a multi-lead residual neural network (ML–ResNet) structure with three residual blocks and feature fusion via 12 leads ECG records. Specifically, single lead feature branch network is trained to automatically learn the representative features of different levels between different layers, which exploits local characteristics of ECG to characterize the spatial information representation. Then all the lead features are fused together as global features. To evaluate the generalization of proposed method and clinical utility, two schemes including the intra-patient scheme and inter-patient scheme are all employed. Results: Experimental results based on PTB (Physikalisch-Technische Bundesanstalt) database shows that our model achieves superior results with the accuracy of 95.49%, the sensitivity of 94.85%, the specificity of 97.37%, and the F1 score of 96.92% for MI detection under the inter-patient scheme compared to the state-of-the-art. By contrast, the accuracy is 99.92% and the F1 score is 99.94% based on 5-fold cross validation under the intra-patient scheme. As for five types of MI location, the proposed method also yields an average accuracy of 99.72% and F1 of 99.67% in the intra-patient scheme. Conclusions: The proposed method for MI detection and location has achieved superior results compared to other detection methods. However, further promotion of the performance based on MI location for the inter-patient scheme still depends significantly on the mass data and the novel model which reflects spatial location information of different leads subtly.