Transcranial magnetic stimulation (TMS) has recently been used as a method for the treatment of neurological and psychiatric diseases. Daily TMS sessions can provide continuous therapeutic effectiveness, and the installation of TMS systems at patients' homes has been proposed. A figure-eight coil, which is normally used for TMS therapy, induces a highly localized electric field; however, it is challenging to achieve accurate coil positioning above the targeted brain area using this coil. In this paper, a bowl-shaped coil for stimulating a localized but wider area of the brain is proposed. The coil's electromagnetic characteristics were analyzed using finite element methods, and the analysis showed that the bowl-shaped coil induced electric fields in a wider area of the brain model than a figure-eight coil. The expanded distribution of the electric field led to greater robustness of the coil to the coil-positioning error. To improve the efficiency of the coil, the relationship between individual coil design parameters and the resulting coil characteristics was numerically analyzed. It was concluded that lengthening the outer spherical radius and narrowing the width of the coil were effective methods for obtaining a more effective and more uniform distribution of the electric field.

There are individual variations on the motor threshold (MT) and therapeutic effect in clinical treatment using transcranial magnetic stimulation (TMS)�These variations may result from the difference of individual brain anatomies. In this study, we built numerical brain models individually from six subjects, and calculated the distributions of eddy currents induced by TMS. The brain models were built from individual MRI data with segmenting into gray matter, white matter, and cerebrospinal fluid. The location of the figure-eight stimulator coil was recorded using a binocular infrared camera when the stimulation response of twitch observed over 50 % of trials. The eddy current distributions were obtained using an originally developed solver based on the scalar potential finite difference (SPFD) method. The results showed different distributions of the eddy current density between each brain models. The average eddy current density in the primary motor cortex was 17±6.9 A/m 2 for the stimulus intensity corresponding to the MT. Assessment of the relationship between the eddy current density, stimulus conditions, and brain anatomy would help understanding of the mechanism of the varying MT. The developed model enabled us to compare the numerical results with experiments. Experiments have shown that a displacement of stimulator coil from the appropriate location causes an increase in the MT. This phenomenon was observed also in our simulations.

No abstract