Fractal dimension as a feature for adaptive electroencephalogram segmentation in epilepsy

In previous studies the fractal dimension (FD) has been shown to be a useful tool to detect non-stationarities and transients in biomedical signals like electroencephalogram (EEG) and electrocardiogram (ECG). The changes in FD are shown to characterise alterations in EEG due to changes in physiological states of brain, not only in normal but also in pathological functioning like epilepsy. The importance of long-term EEG monitoring for clinical evaluation in epilepsy has been also emphasised. Adaptive EEG. segmentation and classification of the obtained segments have been addressed to be a convenient solution to the problem of visual inspection of huge EEG data sets. The performance of adaptive segmentation plays an essential role in correct evaluation of the recordings. Thus, our aim in this study is to analyse the FD as a feature for adaptive EEG segmentation and compare its performance with those of previously used features on epileptic EEG data.