Head Model Simplification Methods Based on Validity Estimate for Transcranial Magnetic Stimulation

The distribution of electric field (E-field) induced by transcranial magnetic stimulation (TMS) can usually obtained by numerical calculation. To improve calculation efficiency, various types of simplified head models were used in previous studies, but no exact research on the influence of different simplification methods on calculation results has been put forward yet. In this article, 56 numerical human head models (including 7 original models) divided into 7 groups were constructed first, then 3 types stimulus coils were performed to get numerical calculation results. The 5th-percentitle, the 10th-percentitle, the 25th-percentitle, the 50th-percentitle, the 75th-percentitle, the 90th-percentitle, the 95th-percentitle and the 99th-percentitle (peak field value) of E-field in gray matter layer of original models and corresponding simplified ones were used to estimate the validity of simplification methods quantitatively. As for simplified model 2, the maximum absolute relative error of percentiles was 6.97%, which was smaller than absolute relative error of other simplified models and meant the simplification method of incorporating the scalp layer into skull layer was applicable to model I~VII. These works could be technical references for further studies which need to use simplified head model in numerical calculation for TMS.