Learning from VAE Errors to support ECG-based Differential Diagnosis of Myocardial Scar
Abstract
Late Gadolinium Enhancement (LGE) on cardiac magnetic resonance is a key marker of myocardial scar, but its limited accessibility motivates routine ECG-based screening. We evaluated whether β-variational autoencoder (VAE)-derived ECG representations can discriminate LGE+ from LGE- cardiomyopathic patients in a local cohort of 300 subjects. We compared 32-dimensional features from the foundation ECGx.AI model with those from a shallower β-VAE trained on normal PTB-XL ECGs, evaluating downstream classification and Dynamic Time Warping (DTW)-based reconstruction errors. ECGx.AI reached an area under ROC of 0.686 with Random Forest, while the proposed β-VAE reached 0.577 with sensitivity of 0.775 with Gradient Boosting. Notably, DTW-reconstruction errors significantly differed between classes in 10 out of 12 leads according to Mann-Whitney U test and help in classification, leading to an area under ROC of 0.643 with Logistic Regression, supporting their potential as markers of scar-related ECG alterations.
Related
- Learning Sparse Support under Differential Privacy: Adaptive Algorithms and Minimax Limits
- Area Measures for Differential Item and Test Functioning on the Logit Scale: Relationships to Linking Errors in the 2PL Model
- New bounds for the support of input-output equations in differential-algebraic systems
- Flowz: When Does Learning Forecast Errors Add Value? Conditional Flow Matching for Financial Price Forecasts
- Learning Coarse Spaces for Two-Grid Methods from Smoothed Errors with Applications to High-Contrast Porous Media Flow
BibTeX
@misc{sharifi2026learning,
title = {{Learning from VAE Errors to support ECG-based Differential Diagnosis of Myocardial Scar}},
author = {Shayan Sharifi and Riccardo Treu and Ilaria Gandin and Federico Garoia and Marco Merlo and Giulia Cisotto},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.05294},
eprint = {2609.05294},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.05294},
}
RIS
TY - UNPB
TI - Learning from VAE Errors to support ECG-based Differential Diagnosis of Myocardial Scar
AU - Shayan Sharifi
AU - Riccardo Treu
AU - Ilaria Gandin
AU - Federico Garoia
AU - Marco Merlo
AU - Giulia Cisotto
PY - 2026
DA - 2026/09/04
PB - arXiv
DO - 10.48550/arxiv.2609.05294
UR - https://arxiv.org/abs/2609.05294
AB - Late Gadolinium Enhancement (LGE) on cardiac magnetic resonance is a key marker of myocardial scar, but its limited accessibility motivates routine ECG-based screening. We evaluated whether β-variational autoencoder (VAE)-derived ECG representations can discriminate LGE+ from LGE- cardiomyopathic patients in a local cohort of 300 subjects. We compared 32-dimensional features from the foundation ECGx.AI model with those from a shallower β-VAE trained on normal PTB-XL ECGs, evaluating downstream classification and Dynamic Time Warping (DTW)-based reconstruction errors. ECGx.AI reached an area under ROC of 0.686 with Random Forest, while the proposed β-VAE reached 0.577 with sensitivity of 0.775 with Gradient Boosting. Notably, DTW-reconstruction errors significantly differed between classes in 10 out of 12 leads according to Mann-Whitney U test and help in classification, leading to an area under ROC of 0.643 with Logistic Regression, supporting their potential as markers of scar-related ECG alterations.
ER -