Electrocardiogram generation with a bidirectional LSTM-CNN generative adversarial network

TLDR

Heart disease poses a major health threat, and ECG tests diagnose it, but automated diagnosis requires large labeled datasets that raise privacy concerns. The study proposes a BiLSTM‑CNN GAN to generate synthetic ECG data that preserve heart‑disease features, addressing data‑privacy constraints. The BiLSTM‑CNN GAN comprises a BiLSTM generator and CNN discriminator, trained on 48 MIT‑BIH ECG records and benchmarked against RNN‑AE and RNN‑VAE models, with discriminator loss evaluated across generator‑discriminator combinations. The BiLSTM‑CNN GAN converged fastest and produced synthetic ECGs with high morphological similarity to real recordings.

Abstract

Abstract Heart disease is a malignant threat to human health. Electrocardiogram (ECG) tests are used to help diagnose heart disease by recording the heart’s activity. However, automated medical-aided diagnosis with computers usually requires a large volume of labeled clinical data without patients' privacy to train the model, which is an empirical problem that still needs to be solved. To address this problem, we propose a generative adversarial network (GAN), which is composed of a bidirectional long short-term memory(LSTM) and convolutional neural network(CNN), referred as BiLSTM-CNN,to generate synthetic ECG data that agree with existing clinical data so that the features of patients with heart disease can be retained. The model includes a generator and a discriminator, where the generator employs the two layers of the BiLSTM networks and the discriminator is based on convolutional neural networks. The 48 ECG records from individuals of the MIT-BIH database were used to train the model. We compared the performance of our model with two other generative models, the recurrent neural network autoencoder(RNN-AE) and the recurrent neural network variational autoencoder (RNN-VAE). The results showed that the loss function of our model converged to zero the fastest. We also evaluated the loss of the discriminator of GANs with different combinations of generator and discriminator. The results indicated that BiLSTM-CNN GAN could generate ECG data with high morphological similarity to real ECG recordings.