Enhancing Multi-Label ECG Classification via Task-Guided Lead Correlations in Internet of Medical Things

Yuan, Xiaoyan, Wang, Wei ORCID: https://orcid.org/0000-0002-1717-5785, Chen, Junxin ORCID: https://orcid.org/0000-0003-4745-8361, Fang, Kai ORCID: https://orcid.org/0000-0003-0419-1468, Bashir, Ali Kashif ORCID: https://orcid.org/0000-0003-2601-9327, Mondal, Tapas ORCID: https://orcid.org/0000-0003-2077-1710, Hu, Xiping ORCID: https://orcid.org/0000-0002-4952-699X and Deen, M. Jamal ORCID: https://orcid.org/0000-0002-6390-0933 (2025) Enhancing Multi-Label ECG Classification via Task-Guided Lead Correlations in Internet of Medical Things. IEEE Internet of Things Journal. ISSN 2372-2541

Preview

Accepted Version Available under License In Copyright. Download (1MB) | Preview

Abstract

With the rise of the Internet of Things (IoT), wearable devices have enabled real-time health monitoring, particularly through physiological signals like electrocardiograms (ECG). The standard 12-lead ECG records the electrical activity of the heart from multiple perspectives, providing valuable insights into cardiac health. However, existing 12-lead ECG analysis methods often treat leads as channel-level arrangements or rely on spatial adjacency to predefine lead connections, limiting their ability to capture the complex spatial and functional relationships between leads fully. To address this limitation, we propose TGLLNet, a task-driven model that automatically learns inter-lead relationships to improve multi-label ECG classification. TGLLNet adaptively learns lead connectivity patterns and relational strengths, enhancing ECG representation and improving model generalizability across tasks. Specifically, TGLLNet employs a Temporal Graph Construction (ETGC) module to convert ECG signals into temporal graphs and uses a Residual Pyramid Graph Convolution (RPG) module for multi-level graph embeddings, utilizing a Graph Convolutional Network (GCN) with independently learnable adjacency matrices. Combined with a Temporal Context Convolution (TCC) module, TGLLNet captures spatio-temporal dependencies, significantly improving ECG representation. Experimental results on 7 tasks from PTB-XL and CPSC2018 datasets demonstrate that TGLLNet outperforms existing methods, showing superior generalizability across different tasks. Our code is available at https://github.com/rosemary333/TGLLnet