EEG signal analysis for seizure detection
Biological tissue obeys Ohm's law for small current densities . Bioimpedance can be measured with four electrodes: two for current injection and two for voltage measurement. The magnitude of the impedance is given by the ratio of the magnitude of the measured voltage and the magnitude of the injected current. There seems to be a slow increase in measured impedance several minutes prior to the onset of seizure activity. This result may be caused by ionic (Ca^^) shifts prior to seizure formulation . However, whether the result is common for most epilepsy patients is still unknown. Besides, there are also some difficulties in determining a suitable warning threshold, and the procedures for bioimpedance measurement are more complicated than that for an EEG . An EEG is the recording of electrical cerebral potentials . AU chemical and physical processes that take place in living cells produce electrical energy. The energy can be recognized by changes in potential of the cell membrane . The potential of one cell is very small, but when a larger group of cells acts simultaneously, the potential is higher and can be recorded with suitable amplification. Many techniques have been developed for EEG signal analysis. Some techniques have been used for seizure detection or prediction, but there have been no conclusive findings. In this thesis, EEG analysis is used as a tool for seizure detection or prediction. Several different signal analysis techniques are examined. The objective is to determine which technique is the best for seizure detection or prediction.