Towards Automated Diagnosis of Inherited Arrhythmias: Combined Arrhythmia Classification Using Lead-Aware Spatial Attention Networks

2026-01-12

arXiv.org

AI Summary

This study addresses the limited interpretability and suboptimal accuracy in automated multi-class electrocardiogram (ECG) diagnosis of inherited arrhythmias, specifically arrhythmogenic right ventricular cardiomyopathy (ARVC) and long QT syndrome (LQTS). To this end, we propose a deep learning framework that integrates a Lead-aware Spatial Attention Network (LASAN) with a pre-trained ECG foundation model such as HuBERT-ECG. By incorporating lead-aware architecture and transfer learning strategies—including fine-tuning and linear probing—the model achieves both interpretable, disease-specific modeling of lead dependencies and near-theoretical performance on multi-class classification. The approach attains a macro-AUROC of 0.990, with binary AUROC scores of 0.999 for ARVC and 0.994 for LQTS, while the derived lead importance aligns closely with established clinical and physiological understanding.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Arrhythmogenic right ventricular cardiomyopathy (ARVC) and long QT syndrome (LQTS) are inherited arrhythmia syndromes associated with sudden cardiac death. Deep learning shows promise for ECG interpretation, but multi-class inherited arrhythmia classification with clinically grounded interpretability remains underdeveloped. Our objective was to develop and validate a lead-aware deep learning framework for multi-class (ARVC vs LQTS vs control) and binary inherited arrhythmia classification, and to determine optimal strategies for integrating ECG foundation models within arrhythmia screening tools. We assembled a 13-center Canadian cohort (645 patients; 1,344 ECGs). We evaluated four ECG foundation models using three transfer learning approaches: linear probing, fine-tuning, and combined strategies. We developed lead-aware spatial attention networks (LASAN) and assessed integration strategies combining LASAN with foundation models. Performance was compared against the established foundation model baselines. Lead-group masking quantified disease-specific lead dependence. Fine-tuning outperformed linear probing and combined strategies across all foundation models (mean macro-AUROC 0.904 vs 0.825). The best lead-aware integrations achieved near-ceiling performance (HuBERT-ECG hybrid: macro-AUROC 0.990; ARVC vs control AUROC 0.999; LQTS vs control AUROC 0.994). Lead masking demonstrated physiologic plausibility: V1-V3 were most critical for ARVC detection (4.54% AUROC reduction), while lateral leads were preferentially important for LQTS (2.60% drop). Lead-aware architectures achieved state-of-the-art performance for inherited arrhythmia classification, outperforming all existing published models on both binary and multi-class tasks while demonstrating clinically aligned lead dependence. These findings support potential utility for automated ECG screening pending validation.

Problem

• inherited arrhythmias
• ECG classification
• ARVC
• LQTS
• automated diagnosis

Innovation

• lead-aware attention
• inherited arrhythmia classification
• ECG foundation models
• spatial attention networks
• transfer learning