Singular spectrum analysis for tracking of P300

Abstract

In this work, we introduce a complex-valued singular spectrum analysis for the analysis of electroencephalogram (EEG), which typically exhibits noncircular probability distribution. To exploit such prior knowledge, our technique makes use of recent advances in complex-valued statistics to exploit the power difference or the correlation between the data channels, in contrast to current methods which cater only for the restrictive class of circular data. In particular, the principal component analysis-like technique was employed to detect the onset of P300, and tracked this event-related potential. In this way, the classification of EEG can be made possible to differentiate between a healthy subject and a schizophrenic patient. In particular, we illuminate how features such as P3a and P3b can be used to perform such classification.

Cooperative Particle Filtering for Tracking ERP Subcomponents from Multichannel EEG

“…Figures 12-17 show the variability of P3a and P3b amplitudes, latencies, and widths over the rare trials for a healthy subject and a schizophrenic patient, respectively. It is observed from these figures that the amplitudes of P300 subcomponents of schizophrenic patients are much smaller compared to healthy subjects, which is in agreement with the existing literature [43][44][45][46]. In fact, one of the most consistent findings in schizophrenic patients is the significant reduction of P300 especially in auditory experiments [43,47].…”

“…In fact, one of the most consistent findings in schizophrenic patients is the significant reduction of P300 especially in auditory experiments [43,47]. Additionally, it can be observed that schizophrenic patients have longer P300 latency compared to healthy subjects, which is in agreement with the results of previous studies [22,43,44]. The mean width of P300 subcomponents is slightly larger for healthy subjects compared to schizophrenic patients.…”

Quaternion Singular Spectrum Analysis of Electroencephalogram With Application in Sleep Analysis

“…However, the advances in multidimensional sensor technologies have highlighted the need for signal processing algorithms in complex and quaternion domain due to their great potential for modelling of twodimensional (2-D) and 4-D data [9]. For instance, the complexvalued SSA enabled the detection of event related potential (ERP) components for the classification of schizophrenic patients [10] and improved forecasting of bivariate signals [11]. In the same spirit, the established common spatial pattern was recently introduced in the complex domain to model the correlation in EEG processing to classify motor imagery tasks [12].…”