f-GAN: A frequency-domain-constrained generative adversarial network for PPG to ECG synthesis

TL;DR

This paper introduces a novel f-GAN model that synthesizes ECG signals from PPG data, improving stability and accuracy in heart rate estimation by incorporating frequency-domain constraints.

Contribution

The study develops a frequency-domain-constrained GAN for PPG to ECG synthesis, enhancing model stability and heart rate estimation accuracy compared to traditional GANs.

Findings

Frequency-domain constraints improve model stability.

Synthesized ECGs enable accurate heart rate extraction.

Model outperforms original GAN formulations.

Abstract

Electrocardiograms (ECGs) and photoplethysmograms (PPGs) are generally used to monitor an individual's cardiovascular health. In clinical settings, ECGs and fingertip PPGs are the main signals used for assessing cardiovascular health, but the equipment necessary for their collection precludes their use in daily monitoring. Although PPGs obtained from wrist-worn devices are susceptible to noise due to motion, they have been widely used to continuously monitor cardiovascular health because of their convenience. Therefore, we would like to combine the ease with which PPGs can be collected with the information that ECGs provide about cardiovascular health by developing models to synthesize ECG signals from paired PPG signals. We tackled this problem using generative adversarial networks (GANs) and found that models trained using the original GAN formulations can be successfully used to synthesize ECG signals from which heart rate can be extracted using standard signal processing pipelines. Incorporating a frequency-domain constraint to model training improved the stability of model performance and also the performance on heart rate estimation.