Comparison of Transcranial Magnetic Stimulation Dosimetry between Structured and Unstructured Grids Using Different Solvers

In recent years, the interest in transcranial magnetic stimulation (TMS) has surged, necessitating deeper understanding, development, and use of low-frequency (LF) numerical dosimetry for TMS studies. While various ad hoc dosimetric models exist, commercial software tools like SimNIBS v4.0 and Sim4Life v7.2.4 are preferred for their user-friendliness and versatility. SimNIBS utilizes unstructured tetrahedral mesh models, while Sim4Life employs voxel-based models on a structured grid, both evaluating induced electric fields using the finite element method (FEM) with different numerical solvers. Past studies primarily focused on uniform exposures and voxelized models, lacking realism. Our study compares these LF solvers across simplified and realistic anatomical models to assess their accuracy in evaluating induced electric fields. We examined three scenarios: a single-shell sphere, a sphere with an orthogonal slab, and a MRI-derived head model. The comparison revealed small discrepancies in induced electric fields, mainly in regions of low field intensity. Overall, the differences were contained (below 2% for spherical models and below 12% for the head model), showcasing the potential of computational tools in advancing exposure assessment required for TMS protocols in different bio-medical applications.

Outcome measures for electric field modeling in tES and TMS: A systematic review and large-scale modeling study

BACKGROUND

Electric field (E-field) modeling is a potent tool to estimate the amount of transcranial magnetic and electrical stimulation (TMS and tES, respectively) that reaches the cortex and to address the variable behavioral effects observed in the field. However, outcome measures used to quantify E-fields vary considerably and a thorough comparison is missing.

OBJECTIVES

This two-part study aimed to examine the different outcome measures used to report on tES and TMS induced E-fields, including volume- and surface-level gray matter, region of interest (ROI), whole brain, geometrical, structural, and percentile-based approaches. The study aimed to guide future research in informed selection of appropriate outcome measures.

METHODS

Three electronic databases were searched for tES and/or TMS studies quantifying E-fields. The identified outcome measures were compared across volume- and surface-level E-field data in ten tES and TMS modalities targeting two common targets in 100 healthy individuals.

RESULTS

In the systematic review, we extracted 308 outcome measures from 202 studies that adopted either a gray matter volume-level (n = 197) or surface-level (n = 111) approach. Volume-level results focused on E-field magnitude, while surface-level data encompassed E-field magnitude (n = 64) and normal/tangential E-field components (n = 47). E-fields were extracted in ROIs, such as brain structures and shapes (spheres, hexahedra and cylinders), or the whole brain. Percentiles or mean values were mostly used to quantify E-fields. Our modeling study, which involved 1,000 E-field models and > 1,000,000 extracted E-field values, revealed that different outcome measures yielded distinct E-field values, analyzed different brain regions, and did not always exhibit strong correlations in the same within-subject E-field model.

CONCLUSIONS

Outcome measure selection significantly impacts the locations and intensities of extracted E-field data in both tES and TMS E-field models. The suitability of different outcome measures depends on the target region, TMS/tES modality, individual anatomy, the analyzed E-field component and the research question. To enhance the quality, rigor, and reproducibility in the E-field modeling domain, we suggest standard reporting practices across studies and provide four recommendations.

Devices and Technology in Transcranial Magnetic Stimulation: A Systematic Review

The technology for transcranial magnetic stimulation (TMS) has significantly changed over the years, with important improvements in the signal generators, the coils, the positioning systems, and the software for modeling, optimization, and therapy planning. In this systematic literature review (SLR), the evolution of each component of TMS technology is presented and analyzed to assess the limitations to overcome. This SLR was carried out following the PRISMA 2020 statement. Published articles of TMS were searched for in four databases (Web of Science, PubMed, Scopus, IEEE). Conference papers and other reviews were excluded. Records were filtered using terms about TMS technology with a semi-automatic software; articles that did not present new technology developments were excluded manually. After this screening, 101 records were included, with 19 articles proposing new stimulator designs (18.8%), 46 presenting or adapting coils (45.5%), 18 proposing systems for coil placement (17.8%), and 43 implementing algorithms for coil optimization (42.6%). The articles were blindly classified by the authors to reduce the risk of bias. However, our results could have been influenced by our research interests, which would affect conclusions for applications in psychiatric and neurological diseases. Our analysis indicates that more emphasis should be placed on optimizing the current technology with a special focus on the experimental validation of models. With this review, we expect to establish the base for future TMS technological developments.

Numerical analysis of transcranial magnetic stimulation application in patients with orofacial pain

In this paper, we monitored the accuracy of non-navigated application of repetitive Transcranial Magnetic Stimulation (rTMS) in 10 patients suffering from orofacial pain by using functional magnetic resonance (fMRI), computer modeling and numerical simulation. Through a unique process, each fMRI scan was used to define a Region of Interest (ROI) where the source of the orofacial pain was located, which was to be stimulated using rTMS. For each patient, MRI scans with a spatial resolution of 0.7 mm were converted into an anatomically accurate head model. The head model including the ROI was then co-registered with a model of the stimulation coil in an electromagnetic field numerical simulator. The accuracy of rTMS application was evaluated based on the calculations of electric field intensity distribution in the ROI. The research has yielded unique insight into ROIs (with average volume 904mm3) in patients with orofacial pain and has also extended further possibilities of human head MRI image semi-automatic segmentation. According to the calculations performed, the average ROI volume that was stimulated by an electric field with an intensity of over 80 V/m was only 4.4%, with the maximum ROI volume being 20.5%. Furthermore, a numerical study of the impact of coil rotation and translation was performed. It demonstrated a) the optimal placement of the stimulation coil can significantly increase the volume of the stimulated ROI up to 60% and b) patients with orofacial pain would need precise coil positioning with a navigation error lower than 10 mm. Due to an acceptable proccessing time of up to 6 hours, described numerical simulation opens up new options for precise rTMS treatment planning. This planning platform together withpatient-specific navigated rTMS, could lead to significant increase of treatment outcomes in patients suffering from orofacial pain.

Integral Equation Formulations and Related Numerical Solution Methods in Some Biomedical Applications of Electromagnetic Fields

The paper reviews certain integral equation approaches and related numerical methods used in studies of biomedical applications of electromagnetic fields pertaining to transcranial magnetic stimulation (TMS) and nerve fiber stimulation. TMS is analyzed by solving the set of coupled surface integral equations (SIEs), while the numerical solution of governing equations is carried out via Method of Moments (MoM) scheme. A myelinated nerve fiber, stimulated by a current source, is represented by a straight thin wire antenna. The model is based on the corresponding homogeneous Pocklington integro-differential equation solved by means of the Galerkin Bubnov Indirect Boundary Element Method (GB-IBEM). Some illustrative numerical results for the TMS induced fields and intracellular current distribution along the myelinated nerve fiber (active and passive), respectively, are presented in the paper.

Neurophysiologic markers of primary motor cortex for laryngeal muscles and premotor cortex in caudal opercular part of inferior frontal gyrus investigated in motor speech disorder: a navigated transcranial magnetic stimulation (TMS) study

Transcranial magnetic stimulation studies have so far reported the results of mapping the primary motor cortex (M1) for hand and tongue muscles in stuttering disorder. This study was designed to evaluate the feasibility of repetitive navigated transcranial magnetic stimulation (rTMS) for locating the M1 for laryngeal muscle and premotor cortical area in the caudal opercular part of inferior frontal gyrus, corresponding to Broca's area in stuttering subjects by applying new methodology for mapping these motor speech areas. Sixteen stuttering and eleven control subjects underwent rTMS motor speech mapping using modified patterned rTMS. The subjects performed visual object naming task during rTMS applied to the (a) left M1 for laryngeal muscles for recording corticobulbar motor-evoked potentials (CoMEP) from cricothyroid muscle and (b) left premotor cortical area in the caudal opercular part of inferior frontal gyrus while recording long latency responses (LLR) from cricothyroid muscle. The latency of CoMEP in control subjects was 11.75 ± 2.07 ms and CoMEP amplitude was 294.47 ± 208.87 µV, and in stuttering subjects CoMEP latency was 12.13 ± 0.75 ms and 504.64 ± 487.93 µV CoMEP amplitude. The latency of LLR in control subjects was 52.8 ± 8.6 ms and 54.95 ± 4.86 in stuttering subjects. No significant differences were found in CoMEP latency, CoMEP amplitude, and LLR latency between stuttering and control-fluent speakers. These results indicate there are probably no differences in stuttering compared to controls in functional anatomy of the pathway used for transmission of information from premotor cortex to the M1 cortices for laryngeal muscle representation and from there via corticobulbar tract to laryngeal muscles.