A Usable GAN-Based Tool for Synthetic ECG Generation in Cardiac Amyloidosis Research
This work addresses data scarcity in a rare disease context, but it is incremental as it applies an existing GAN method to a new medical domain.
The paper tackled the problem of small, imbalanced datasets for cardiac amyloidosis research by developing a GAN-based tool to generate realistic synthetic ECG beats, enabling early diagnosis and patient stratification with usability for clinical researchers.
Cardiac amyloidosis (CA) is a rare and underdiagnosed infiltrative cardiomyopathy, and available datasets for machine-learning models are typically small, imbalanced and heterogeneous. This paper presents a Generative Adversarial Network (GAN) and a graphical command-line interface for generating realistic synthetic electrocardiogram (ECG) beats to support early diagnosis and patient stratification in CA. The tool is designed for usability, allowing clinical researchers to train class-specific generators once and then interactively produce large volumes of labelled synthetic beats that preserve the distribution of minority classes.