Synthesis of standard 12-lead electrocardiograms using two dimensional generative adversarial network

TL;DR

This paper introduces a 2D GAN model capable of generating realistic synthetic 12-lead ECG signals for various cardiac conditions, aiding in expanding ECG datasets for research and clinical purposes.

Contribution

The study presents a novel 2D bidirectional LSTM GAN that automatically generates diverse, unbiased synthetic 12-lead ECGs for multiple cardiac conditions, with high success rates.

Findings

Success rates: 98% for LVH, 93% for LBBB, 79% for ACUTMI, 59% for Normal.

Synthetic ECGs are statistically similar to real data and cover diverse morphological features.

The model can augment ECG datasets, addressing data scarcity.

Abstract

This paper proposes a two-dimensional (2D) bidirectional long short-term memory generative adversarial network (GAN) to produce synthetic standard 12-lead ECGs corresponding to four types of signals: left ventricular hypertrophy (LVH), left branch bundle block (LBBB), acute myocardial infarction (ACUTMI), and Normal. It uses a fully automatic end-to-end process to generate and verify the synthetic ECGs that does not require any visual inspection. The proposed model is able to produce synthetic standard 12-lead ECG signals with success rates of 98% for LVH, 93% for LBBB, 79% for ACUTMI, and 59% for Normal. Statistical evaluation of the data confirms that the synthetic ECGs are not biased towards or overfitted to the training ECGs, and span a wide range of morphological features. This study demonstrates that it is feasible to use a 2D GAN to produce standard 12-lead ECGs suitable to augment artificially a diverse database of real ECGs, thus providing a possible solution to the demand for extensive ECG datasets.