Grant: Collaborative Health Research Projects (NSERC partnered)

T-wave alternans (TWA) refers to the changes in T wave shape or amplitude appearing every second heart beat on a surface electrocardiogram (ECG). The magnitude of TWA signal is very small, in the microvolt range, but some clinical trials have recently suggested that its presence can predict sudden death in patients with heart disease. From a signal analysis perspective, the accurate measurement of the TWA signal is challenging due to very low signal to noise ratio (SNR), and signal non-stationarity. Several noise sources can potentially confound TWA signal detection, including periodic noise from movement or respiration, and non-periodic/spurious noise. The primary objective of this proposal will be to develop a more accurate TWA detection algorithm that tackles the signal non-stationarity and provides robust measurement of TWA signal parameters even in low SNR conditions.In the proposed techniques the non-stationarity of the TWA signal is handled by transforming the signal to a joint time-frequency representation (TFR) which provides the time-varying spectral representation. The marginals computed as first-order and second-order moments along time or frequency direction in a proper TFR will provide features such as instantaneous frequency (IF) and instantaneous bandwidth (IBW). IF and IBW could be used to characterize the non-stationary TWA and thereby lead to better detection and estimation of TWA. The secondary objective will be to evaluate the accuracy of the proposed techniques in risk stratifying patients with heart disease. This will be achieved by measuring TWA every 6 months from a 24-hour ambulatory ECG recording in 450 patients with heart disease, and then following these patients prospectively for 2 years to determine whether TWA predicts subsequent death or nonfatal ventricular arrhythmias. Developing accurate risk assessment algorithms is essential to identify patients at highest risk of sudden death who will derive the greatest survival benefit from costly medical device therapy, such as implantable defibrillators.