Signal-to-noise ratio decision trees for extracting continuous, reliable and real-time heart rate variability (HRV) measurements from an ambulatory ECG monitoring device

点击翻译按钮获取中文摘要

The goal of this research project is to develop computational algorithms for processing ambulatoryElectrocardiogram (ECG) signals recorded in real time by a wearable device and accurately measuring thevariation in the time intervals between each heartbeat, referred to as heart rate variability (HRV). HRV is animportant vital signal used by physicians and other healthcare professionals to monitor and assess the risk ofseveral cardiac and non-cardiac disorders. The main technical challenges this research sets out to overcomeinvolve reliably extracting HRV measurements despite common disturbances present in ambulatory ECGmonitoring and efficiently designing the algorithms to operate in real-time to allow for continuousmeasurements of the subject's current HRV. The partner organization intends to add the developed algorithmsto their existing remote monitoring device, which is currently capable of recording ECG amongst otherphysiological and physical activity signals. By adding the capability of accurate HRV measurement to thewearable device and monitoring platform developed by the industry partner, a crucial vital sign will beincorporated, thus further facilitating the adaptation of remote monitoring systems by the Canadian HealthcareSystem. Individuals known to be at risk of specific medical conditions and hospital patients recovering fromoperations where vital sign monitoring is needed can benefit from remote monitoring in the comfort of theirhomes. This will in turn reduce the burden incurred to hospitals and healthcare professionals and greatly reducethe associated hospitalization costs.