An investigation of the specificity of features of early stages of Parkinson's disease obtained using the method of cortex electrical activity analysis based on wave trains

We developed a new method of signal analysis based on wavelet analysis, ROC-analysis, and non-parametric statistics for detailed investigation of the time-frequency dynamics of the electrical activity of the cerebral cortex. The idea of the method is in that the electroencephalogram (EEG) is considered as a set of wave trains (WT). WT is detected as a local maximum in the wavelet spectrogram of EEG. We consider WT as a typical component of EEG, but not as a special kind of EEG signals. The following parameters of WT are accounted: the frequency, the duration, the bandwidth, the number of WT per second, and the power spectral density (PSD). Differences between a group of the first stage Parkinson's disease patients and a group of healthy volunteers in the space of these parameters are investigated. ROC-analysis is used for this purpose. We analyzed functional dependence of AUC on the boundaries of the ranges of these parameters. Using this method, we have identified three frequency ranges, where differences between the group of the patients and the healthy volunteers were discovered. The paper describes the results of the investigation of the specificity of these features of early stages of Parkinson's disease.