Extensive deep learning model to enhance electrocardiogram application via latent cardiovascular feature extraction from identity identification. (April 2023)

• Record Type: Journal Article
• Authors: Lou, Yu-Sheng; Lin, Chin-Sheng; Fang, Wen-Hui; Lee, Chia-Cheng; Lin, Chin
• Is Part Of: Computer methods and programs in biomedicine. Volume 231(2023)
• Publication Date: 2023-04
• Subjects: Electrocardiogram -- Deep learning -- Cardiovascular disease -- Transfer learning -- Unsupervised learning -- Identity identification; 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.2023.107359
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: A novel transfer learning strategy based on identity identification for extracting cardiovascular-related features from an electrocardiogram is proposed. We pre-train an ECG-based deep learning model with identity identification and fine-tune it for predicting 70 patient characteristics. By using transfer learning based on identity identification, we significantly improve the performance for diagnosing related diseases and predicting future cardiovascular diseases. Abstract: Background and Objective: Deep learning models (DLMs) have been successfully applied in biomedicine primarily using supervised learning with large, annotated databases. However, scarce training resources limit the potential of DLMs for electrocardiogram (ECG) analysis. Methods: We have developed a novel pre-training strategy for unsupervised identity identification with an area under the receiver operating characteristic curve (AUC) >0.98. Accordingly, a DLM pre-trained with identity identification can be applied to 70 patient characteristic predictions using transfer learning (TL). These ECG-based patient characteristics were then used for cardiovascular disease (CVD) risk prediction. The DLMs were trained using 507, 729 ECGs from 222, 473 patients and validated using two independent validation sets (n = 27, 824/31, 925). Results: The DLMs using our method exhibited better performance than directly trained DLMs. Additionally, our DLM performed better than those of previous studies in terms of gender (AUC [internal/external] = 0.982/0.968), age (correlation = 0.886/0.892), low ejection fraction (AUC = 0.942/0.951), and critical markers not addressed previously, including high B-type natriuretic peptide (AUC = 0.921/0.899). Additionally, approximately 50% of the ECG-based characteristics provided significantly more prediction information for cardiovascular risk than real characteristics. Conclusions: This is the first study to use identity identification as a pre-training task for TL in ECG analysis. An extensive exploration of the relationship between ECG and 70 patient characteristics was conducted. Our DLM-enhanced ECG interpretation system extensively advanced ECG-related patient characteristic prediction and mortality risk management for cardiovascular diseases. Graphical abstract: Image, graphical abstract