The preliminary study of EEG and ECG for epileptic seizure prediction based on Hilbert Huang Transform

Epilepsy is a chronic brain disorder. The patient are suffer from the unpredictable seizure. The conventional method for studies the characteristic of epileptic seizure is measuring the Electroencephalogram (EEG). On the other side, there are some studies reported about the relation between heart rate from Electrocardiogram (ECG) and epileptic seizure. This paper is a preliminary study about EEG and ECG based epileptic seizure prediction. The feature extraction method is based on the Hilbert Huang Transform (HHT) and we try to indicate some phenomena of EEG and ECG before the seizure onset. We extract the mean instantaneous frequency from EEG and R-R interval from ECG. The result shows the mean instantaneous frequency in mode one of intrinsic mode function was significantly dropped down simultaneously with R-R interval variation before seizure onset, the prior time is around 130 second. So, we can conclude that there is a possibility to use these two feature as a indicator for early prediction.