Non-invasive Functional Neuroimaging for Localizing Epileptic Brain Activity

Licentiate thesis, 2012

Surgical therapy has become an important therapeutic alternative for patients with medically intractable epilepsy. Correct and anatomically precise localization of the epileptic focus, preferably with non-invasive methods, is the main goal of the pre-surgical epilepsy diagnosis to decide if resection of brain tissue is possible. The most important diagnosis tool used at epilepsy surgery centers is electroencephalography (EEG), which is used to find the source of activities inside the brain by measuring the voltage potential on the scalp with the EEG electrodes at different locations. The overall goal is to develop a non-invasive, clinically-viable, time-efficient method for localization of epileptic brain activity based on EEG source localization. We propose a new global optimization method based on particle swarm optimization (PSO) to solve the epileptic spike EEG source localization inverse problem. In the forward problem a modified subtraction method is used for modeling the dipole source to reduce the computational time. The new proposed inverse method is tested for synthetic and real EEG data and the results are compared with other existing methods. The results for synthetic data showed that the new PSO algorithm can find the optimal solution significantly faster and more accurate than the other methods and also reduce the probability of trapping in local minima. In the clinical test, somatosensory evoked potentials (SEPs) were measured for a healthy subject and used for source localization. A realistic 1 mm patient-specific, isotropic finite element model of the subject’s head with special consideration of precise modeling the two compartments, skull and cerebrospinal fluid (CSF), was generated using T1-weighted magnetic resonance imaging data. The proposed inverse problem solver found the global minima with acceptable accuracy and reasonable number of iterations.