Quantification of brain injury by EEG cepstral distance during transient global ischemia

Clinical assessment of neurological system would be greatly facilitated by interpretation of disease- or injury-related changes in electroencephalogram (EEG) signals. Autoregressive (AR) models have previously been used to parameterize the EEG signals. Cepstral distance which is based on the AR model parameters, is used here to measure the similarity of the EEG signals obtained during various phases of the experimental study. The effectiveness of the cepstral distance is demonstrated through its ability to distinguish asphyxia and quantify the period of the brain asphyxia injury