Year: 2025
Type: article
Abstract: Artificial intelligence-enhanced electrocardiogram (AI-ECG) models have shown outstanding performance in diagnostic and prognostic tasks, yet their black-box nature hampers clinical adoption. Meanwhile, a growing demand for explainable AI in medicine underscores the need for transparent, trustworthy decision-making. Moving beyond post-hoc explainability techniques that have shown unreliable results, we focus on explicit representation learning using variational autoencoders (VAE) to capture inherently interpretable ECG features. While VAEs have demonstrated potential for ECG interpretability, the presumed performance-explainability trade-off remains underexplored, with many studies relying on complex, non-linear methods that obscure the morphological information of the features. In this work, we present a novel framework (VAE-SCAN) to model bi-directional, interpretable associations between ECG features and clinical factors. We also investigate how different representations affect ECG decoding performance across models with varying levels of explainability. Our findings demonstrate the cost introduced by intrinsic ECG interpretability, based on which we discuss potential implications and directions.
Source: npj Digital Medicine
Institutions: Imperial College London, University of Antwerp, Antwerp University Hospital, Universidade Federal de Minas Gerais, Uppsala University, Hospital das Clínicas da Universidade Federal de Minas Gerais, Imperial College Healthcare NHS Trust, Guy's and St Thomas' NHS Foundation Trust, Harefield Hospital, Royal Brompton Hospital, Beth Israel Deaconess Medical Center, Harvard University, Chelsea and Westminster Hospital NHS Foundation Trust, Chelsea and Westminster Hospital
Language: English
Topic: ECG Monitoring and Analysis
Subfield: Cardiology and Cardiovascular Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $4,090
Funder: British Heart Foundation
OpenAlex ID: W4417034174