Dimensional complexity of the EEG in subcortical stroke--a case study

The conventional electrophysiological methods used for the analysis of the functional characteristics of the nervous system are not able to grasp its non-linear and random features. Of the methods based on the application of chaos-theory the correlation dimension analysis can be used to quantify the complexity of the analyzed signal, such as the electroencephalogram (EEG). The new version (point-correlation dimension, PD2) was used in this study, which is more accurate than the other, currently used algorithms. The purpose of the present investigation was to compare the sensitivity of the methods based on chaos-theory with the traditional electrophysiological ones in a case when no apparent abnormality was present as judged on the basis of this latter methodology. The PD2 was calculated from the EEG recorded in 13 healthy control subjects and in a patient who suffered a small subcortical stroke 2 years prior to the investigation and who was free of neurological symptoms at the time of recording. Compared to that seen in the control group, in the Z-score maps of the scalp distribution of the PD2, a marked asymmetry was seen and the absolute PD2 values showed a low-dimensional area in the parietal region, ipsilateral to the stroke. A relative decrease of the gamma band was found in the frequency power spectra in the same area. It is suggested that the additional information extracted from the EEG by non-linear analysis may increase the sensitivity of electrophysiological methods for detecting brain pathology.