Multiresolution Characteristics And Its Clinical Application Of Electroencephalograph

Objects:To study the transient multi-scale features and multi-scale power distribution of the digital EEG signal and compare multi-scale features and multi-scale powers of EEG signal across scales in normal subjects and two kinds of disease. Capture the respective multi-scale features in different kinds of subjects. Extract the evolution mechanism of multi-scale features in epileptic discharge. Explore a new tool for digital EEG analysis, which is helpful to clinical diagnosis and basis research of neurology.Methods:(1) Analysis the digital EEG signals of 20 normal adults at awake and eye-closed state with multi-scale resolution by wavelet transform. Extract the qualitative multi-scale features, power distributions across frequency and coordination of scalp.(2) Analysis the digital EEG signals of 22 children at awake and eye-closed state with multi-scale resolution by wavelet transform. Extract the qualitative multi-scale features, power distributions across frequency and coordination of scalp. Capture the evolution of multi-scale features with age increasing by comparing with adults.(3) Analysis the digital EEG signals of 10 absence seizure and 20 subclinical epileptic discharge in 15 CAE patients with multi-scale resolution by wavelet transform. Extract the qualitative features of the seizure onset in multi-scale compared with that of before and after the seizure onset, as well as compared with digital EEG signals of the normal children at the same age.(4) Analysis the digital EEG signals of 30 GTS patients with multi-scale resolution by wavelet transform. Extract the qualitative multi-scale features, power distribution across frequency. Capture multi-scale features by comparing with subjects at same age.Results:(1) The multi-scale characteristic feature of the normal adults is that the wavelet coefficient is relatively small, its frequency range is relatively wide, the activity rhythm is relatively strong and relatively stable in several special scale, the correlation between adjoining scale is intimate, three power peak locate in 0.1 Hz> lHz, lOHz.(2) Compared with normal adults, the multi-scale characteristic feature of the children is that the wavelet coefficient is relatively large, the activity rhythm is relatively unsteady in corresponding scale, the correlation between adjoining scale is less intimate than adults, signal power peak locate in scale 15. Its multi-scale characteristic features are tending towards adults with age increasing.(3) The multi-scale characteristic feature of the absence seizure is the activity rhythm in the scale 10-14 increases, especially in scale 12. Their EEG signals 10s before and after onsets are different from normal. The main representation is the multi-scale power of the seizure onset is mainly localized in scale 20 and scale 12. The power in scale 20 decreases gradually, at the same time the power in scale 12 increase gradually. On space, the power in scale 20 is mainly localized in occipital region as well as the power in scale 12 is localized in frontal and parietal region at seizure start, and shift to occipital region.(4) The multi-scale characteristic feature of the GTS is that the wavelet coefficient is relatively large as compared with contrast at the same age, but is relatively small compared with epileptic discharge, the activity rhythm is undulate, but don't appear the super strong activity rhythm in mono scale. Its multi-scale characteristic features are similar as children compared with adults. It reflects that some EEG features of GTS grow slowly than same age contrast, but it is different with epileptic discharge.