Features extraction of EEG signals using approximate and sample entropy

There are numerous types of mental and neurological disorder where the electroencephalogram (EEG) data size is too long and requires a long time to observe the data by clinician. EEG waveform may contain valuable and useful information about the different states of the brain. Since biological signal is highly random in both time and frequency domain. Thus the computerized analysis is necessary. Being a non-stationary signal, suitable analysis is essential for EEG to differentiate the normal/epileptic and alcoholic/control EEG signals. Approximate entropy (ApEn) and Sample entropy (SampEn) are used to take out the quantitative entropy features from the different types of EEG time series. Average value of ApEn and SampEn for epileptic time series is less than non epileptic time series. Similarly ApEn and SampEn values for alcoholic EEG time series is less than non-alcoholic or control EEG signal.