A novel approach to localize cortical TMS effects

Abstract

21 22 1 METHODS 24 1.1 Determining the site of stimulation 25In the following, the two approaches to determine the congruence factor are presented. The site of effective stimulation 26 is determined by combining the measured I/O-curves of the MEP amplitude with the simulation results of the induced 27 electric field. An operator is derived to transform the brain-wide MEP amplitude curves with 1, … , 28indicating the experimental condition, given as a function of stimulator intensities x, to an element-wise quantity , 29given in the electric field space, i.e. . The quantity is termed the congruence factor. The basic 30 principle to determine is shown in Fig. 5. For each experimental condition, the corresponding electric field 31 distribution is determined. Due to the linear relationship between the stimulator intensity and the magnitude of the 32 induced electric field, the MEP curves can be projected and accordingly scaled to each element in the cortex. As a result, 33the global stimulator intensity vs MEP curves are transformed from the stimulator intensity space to the 34 electric field space , containing spatial information. Therefore, a series of MEP curves is assigned to 35 each element. The definition of the congruence factor is based on the assumption that the causative electric field, which 36 stimulates the neuronal population of interest resulting in the observed behavioral effect is stable across the 37 experimental conditions. For this reason, we are seeking the cortical region, where all MEP curves are similar and 38 assume that this region corresponds to the site of stimulation. This principle is illustrated in Fig. 6, highlighting an 39 element with a low congruence factor, where the E-MEP curves are strongly diverging from each other and an element 40 with a high congruence factor, where the E-MEP curves are similar. The congruence factor is computed in the ROI 41considering the magnitude of electric field | | as well as its normal (| |) and tangential (| ∥ |) components. Two 42 different approaches can be applied to determine the congruence factor, which are described in the following. 43 1.2 Parametrizing the fractional anisotropy of the electrical conductivity tensors 44 Here, we present the methodology to vary the level of anisotropy of the electrical conductivity of the head model, which 45 affects the induced electric field profile. The anisotropic properties of GM and WM result from the alignment of the 46 pyramidal cells in the cortex and the fiber pathways in the white matter 1 . The level of anisotropy is quantified using the each iteration. In order to decrease the computational cost substantially, the electric fields are estimated based on a 112 separate gPC-approximation. The electric field gPCs are determined with the adaptive algorithm proposed by Saturnino 113 et al., (2018) 3 in a pre-processing step with a leave-one-out cross validation error of <0.05%. The electric field gPC 114 approximations act as sub-modules in the model of the congruence factor. Depending on the experimenta...

Multimodal finger-printing of the human precentral cortex forming the motor hand knob

“…This also applies to the precentral gyrus hosting the motor hand representation. The preferential site of direct TMS-induced excitation is the crown-and-lip region, because the electrical field strength is highest in this most superficial part of the precentral gyrus (8)(9)(10). Cytoarchitectonic mapping studies showed that the rostral border of the primary motor cortex is located at the surface of the precentral crown close to the midline, but disappears in the rostral bank of the central sulcus in more lateral parts of the hemisphere (11,12).…”

Review on biophysical modelling and simulation studies for transcranial magnetic stimulation

“…The individually calculated mean EF strength in the motor cortex significantly correlated with the measured RMT (R 2 = 0.44). Weise et al (2020) performed motor-cortical-based TMS measurements using several coil locations and orientations in 15 subjects and modelled the induced EFs using MRIbased head models. By investigating the congruence of the calculated EF and the measured MEP amplitudes, the authors showed that the origin of MEPs was around the gyral crowns and upper parts of the sulcal wall, and that the EF strength was the most relevant quantity to explain the observed effects.…”

“…For the past 30 years, computational dosimetry has progressed significantly in biophysical and electrophysiological modelling techniques to investigate the effects of electromagnetic fields in the human body. In TMS, the primary physical agent known to activate neurons is the induced electric field (EF); moreover, recent studies have enabled the prediction of the stimulation location and optimised the dosages of the stimulation parameters (Opitz et al, 2014;Aonuma et al, 2018;Seynaeve et al, 2019;Weise et al, 2020). These procedures estimate the EF in the brain while considering the effects of the various physical aspects involved in TMS (e.g.…”

Fast computational optimization of TMS coil placement for individualized electric field targeting

“…We compared results obtained by directly computing 〈‖ ( )‖〉 with ones for ‖〈 ( )〉‖ for ROIs of varying sizes to assess the accuracy of this approximation. Furthermore, we ran SimNIBS [47] coil placement optimization to maximize 〈‖ ( )‖〉 and compared with our results obtained by maximizing ‖〈 ( )〉‖. This was accomplished within SimNIBS by co-registering the diffusion-weighted imaging data employing a volume normalization of the brain conductivity tensor, where the geometric mean of the tensor eigenvalues was set to the default isotropic conductivities of 0.275 and 0.126 S/m for gray and white matter, respectively.…”