Abstracts

Rationale: We investigate the feasibility of noninvasive focal stimulation of brain tissue through simulation studies by examining the effects of small amounts of injected currents from scalp electrodes on an anatomically realistic head and brain model. Methods: We constructed a realistic finite element model of the human head built from segmented MR images. The model included 11 different tissue-types, including, CSF, gray and white brain matter, soft and hard skull bone, muscle, fat, scalp etc. The model resolution was 2x2x3.2 mm and the model extended from the top of the head to the cervical-dorsal junction. We simulated a 9x9 scalp grid electrode array over the left temporal area with an interelectrode spacing of 2.0 cm. The central electrode was set at -1.0 volt while the outer electrodes were set at 1 volt. The flux densities were computed in the whole head model with an adaptive finite element solver. The published tissue conductivity values were used for simulation studies. For a comparative analysis, a two electrode model over the left temporal area was also simulated. Electrodes were placed 6 cm apart and had 1 volt on one electrode and -1 on the other electrode. The flux densities were plotted and analyzed in various tissues of each slices. Results: For the grid electrode arrangement, the majority of the currents were confined in an area slightly larger than the electrode grid size. This was observed from the scalp up to a depth of 3 cm in the head where the electrode grid was located. An average volume current density of about 5 microamps/(cm3) was observed in the brain tissue. For the two electrode system, the currents were spread in a large volume of the brain tissue and were not as focused as compared to the grid electrode system. Conclusions: These preliminary results demonstrate the feasibility of selectively localizing stimulating currents to a restricted volume of the brain based on a careful consideration of the size and configuration of stimulating scalp electrode arrays. The amount of stimulating current is similar to that injected routinely for impedance measurements during routine standard EEG recordings.