A Novel Approach for Extracting Nociceptive-Related Time-Frequency Features in Event-Related Potentials

Abstract

In the present study, we are mainly objective to develop a novel approach to extract nociceptive-related features in the time-frequency domain from the event-related potentials (ERPs) that were recorded using high-density EEG. First, the independent component analysis (ICA) was used to separate single-trial ERPs into a set of independent components (ICs), which were then clustered into three groups (symmetrically distributed ICs, non-symmetrically distributed ICs, and noise-related ICs). Second, the time-frequency distributions of each clustered group were calculated using continuous wavelet transform (CWT). Third, the principal component analysis (PCA) with varimax rotation was used to extract time-frequency features from all single-trial time-frequency distributions across all channels. Altogether, the developed approach would help effectively extracting nociceptive-related time-frequency features, thus yielding to an important contribution to the study of nociceptive-specific neural activities.