Cardiac Time Intervals under Motion Using Bimodal Chest E-Tattoos and Multistage Processing.

Abstract

We present a thin, soft, and chest conformable bimodal sensor, known as the chest e-tattoo, coupled with an advanced signal processing framework to accurately identify various cardiac events, and thereby extract cardiac time intervals even during body motion. We built a wireless electronic tattoo that features synchronous electrocardiography (ECG) and seismocardiography (SCG). The SCG measures chest vibrations due to heartbeat, providing information on cardiovascular health that is complementary to the ECG. However, the efficacy of the SCG is compromised by motion-induced artifacts. The slim and stretchy design of the e-tattoo allows it to be strategically placed near the xiphoid process, facilitating high-quality monitoring of the ECG and SCG for increased signal quality. Nine participants were measured during walking and cycling. We propose a multistage signal processing framework, integrating an adaptive Normalized Least Mean Squares (NLMS) filter, ensemble averaging, and Empirical Mode Decomposition (EMD), together named the FAD framework, to accurately extract cardiac time intervals (CTIs).