A Comparison of Time Delay Estimation Methods and Interpolation Methods in Signal-Averaged ECG: Preliminary Results

Signal-Averaged Electrocardiogram (SAECG) is a technique used to analyze ECG records for the detection of subtle changes in the electrical activity of the heart. The alignment (or time delay estimation) of PQRST complexes is the core process in creating an SAECG. In this work, we employed several time delay estimation and interpolation methods to produce SAECG using 60 normal ECG records with added synthesized -30 dB white noise. Performance evaluation was carried out by calculating the ratio of signal to noise ratio (SNR) before and after signal averaging. Spline interpolation method consistently resulted in better performance compared to the linear and cosine methods. Additionally, the results of this study indicate that the sum of square differences (SSD) and the sum of absolute differences (SAD) methods have comparable performances to that of the computationally expensive cross-correlation method.