Combined matching pursuit and Wigner-Ville Distribution analysis for the discrimination of ictal heart rate variability

Abstract

This paper presents a novel method for the discrimination of ictal heart rate variability (HRV). Traditionally, the analysis of the non-linear and non-stationary electrocardiogram (ECG) signal is limited to the time-domain or frequency-domain. This severely limits the quality of features that can be extracted from the ECG signal. In this work, HRV extracted from ECG is analyzed by combining the Matching-Pursuit (MP) and Wigner-Ville Distribution (WVD) algorithms in order to obtain a high quality time-frequency distribution of the HRV signal and to effectively extract meaningful HRV features representative of seizure and non-seizure states. The proposed method is tested on clinical patients and the results demonstrate effective discrimination between ictal HRV features and non-ictal HRV features.