Abstract

The method of analysis of electroencephalograms (EEG) on the basis of wavelet transformations is offered. Electroencephalogram (EEG) analysis is widely used in clinical practice for diagnosing such neurological diseases as epilepsy, Parkinson's disease and others. Traditional approaches to EEG analysis, generally accepted in the clinical diagnosis of diseases, are due to the fact that for a certain time after the stimulus, the EEG amplitudes are calculated at time intervals that depend on the frequency of signal quantization. Therefore, it is important to develop algorithms for classifying EEG signals using wavelet transforms. The analysis of peak-wave EEG discharges, which are indicators of the presence or absence of absence epilepsy, was performed. The EEG recording areas were decomposed into the main EEG frequency bands. Wavelet transform in combination with artificial neural networks makes it possible to implement a classifier based on the energy distribution of the components of the EEG signal. Determining the activity of individual components of EEG signals, as well as the materiality of the processes that take place in the sources of these waves, may be the subject of further research.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call