The Effect of Coil-to-Cortex Distance and Induced Electric Field Strength on Resting Motor Thresholds in Schizophrenia Patients

Transcranial magnetic stimulation (TMS) is a noninvasive neuromodulation technique utilized to treat several neuropsychiatric disorders. Schizophrenia patients typically experience positive and negative symptoms, which are commonly treated with antipsychotic medications. Repetitive TMS (rTMS) is explored to alleviate these symptoms by modulating cortical excitability. TMS stimulation strength is determined by the resting motor threshold (RMT), which is determined from single pulse stimulation to the precentral gyrus (PreCG) region of the brain. In this study, 62 schizophrenia and 41 healthy participants’ magnetic resonance imaging (MRI) were used to create anatomically accurate head models. Head models were then used to calculate the coil-to-cortex distance (CCD) and gray matter volume (GMV), and finite element analysis (FEA) was performed to calculate the induced electric field strength (EFS) and depolarized GMV (DGMV). These MRI-derived parameters were then evaluated to establish their relationship with clinically collected RMT in both healthy and schizophrenia populations. In both populations, RMT correlated with CCD, EFS, and DGMV. Our hypothesis states that the presence of schizophrenia would have a negative impact on the influence of MRI model-driven parameters was largely supported. Future investigations into RMT variability should incorporate structural connectivity using white matter-based fiber tractography and functional connectivity.

The Effect of Coil-to-Cortex Distance and Induced Electric Field Strength on Resting Motor Thresholds in Schizophrenia Patients

I. INTRODUCTION
S CHIZOPHRENIA is a psychotic disorder with a global prevalence of around 1% and is characterized by the presence of positive and/or negative symptoms [1], [2].Positive symptoms include auditory or visual hallucinations and delusions, while negative symptoms include social Manuscript received 28  withdrawal, apathy, and reduced emotional expression [2], [3].These symptoms are typically treated with antipsychotic medications, which act as partial dopamine agonists or dopamine antagonists of dopamine receptors in the brain.Anti-dopaminergic medications are effective in addressing positive symptoms of schizophrenia for approximately 50% of patients, with the others being unresponsive to medication [4].
In addition, antipsychotic drugs may have little impact on many of the negative and cognitive symptoms [2].TMS is a noninvasive neuromodulation technique Food and Drug Administration (FDA)-approved to treat depression, migraine, smoking cessation, and obsessive-compulsive disorder [5], [6], [7], [8].rTMS has previously shown effectiveness in treating neurological conditions by influencing site-specific cortical excitability [9], [10].Thus, rTMS has the potential to treat hallucinating patients with schizophrenia who are refractory to antipsychotic drugs [11], [12].While 1 Hz rTMS to the left temporoparietal cortex appears to treat auditory hallucinations [2], higher frequency rTMS (10 Hz) at the left dorsolateral prefrontal cortex (DLPFC) has been used to treat persistent negative symptoms and cognitive defects by increasing excitability in this region [2], [13], [14].
To determine the dosage stimulation strength of TMS, the RMT for each individual patient must first be established.The RMT is defined as the minimum stimulation intensity that produces a motor-evoked potential (MEP) >50 µV (or a visible motor twitch) in >five out of ten trials in a relaxed muscle on the contralateral side [15], [16], [17].Generally, a proportion (80%-120%) of this measured RMT strength 0018-9464 © 2023 IEEE.Personal use is permitted, but republication/redistribution requires IEEE permission.
Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.
is then applied as the stimulation dose to the region of interest (ROI), such as the DLPFC or primary auditory cortex (PAC) [18], [19].Thus, RMT is a critical patient characteristic in TMS therapy, and it is important to fully understand its variability.
Variability in inter-and intra-patient RMT measurements has previously been shown in multiple TMS participant populations, including those with brain tumors, depression, traumatic brain injuries, and in healthy populations [20], [21], [22], [23], [24], [25].Previous studies have considered factors impacting the RMT or TMS treatment for symptoms of schizophrenia and other neuropsychiatric disorders [26], [27].However, we are not aware of any studies which have individualized finite element models from MRI to investigate the relationships between RMT and model-derived parameters in schizophrenia patients.The aim of this pilot study was to evaluate the relationships between RMT and MRI model-derived parameters, including GMV, age, CCD, induced EFS, and DGMV [20], [21], [22], in comparison between schizophrenia and healthy participants.Our central hypothesis was that the presence of schizophrenia, a brain disorder with widespread abnormalities, will reduce the influence of these parameters on RMT.

A. Participants and TMS Sessions
Participants were recruited from the outpatient services of the National Institute of Mental and Neurosciences (NIMHANS).Data were obtained from an investigational TMS study investigating antipsychotic Naïve (never treated) patients with schizophrenia who were consulting for the first time.The study included 62 schizophrenia participants [age: 34.29 ± 11.0 year; 36(58%) females; and 2.75 ± 2.36 year illness duration] and 41 healthy participants [age 31.78 ± 9.67 year and 21(51%) females] all of whom were included in this study.Diagnosis of schizophrenia was made based on the Diagnostic and Statistical Manual of Mental Health Disorder (DSM-IV-TR) criteria [27] by a qualified psychiatrist, who confirmed using the Mini International Neuropsychiatric Interview [28].Symptom severity was assessed in the schizophrenia population using the PANSS (126.34 ± 16.5) [29].
TMS sessions were conducted using a MagVenture MCF-B70 figure-of-eight butterfly coil (MagVenture, Farum, Denmark).M1 region localization was done using a trial-anderror method by stimulating the left motor cortex about 5 cm lateral to the vertex and marked on a patient cap [30].Directly afterward, TMS pulses were administered in increasing increments of 2% of the coil output to approximate the RMT, and 1% adjustments were made to finalize the measurement.RMT values were recorded as the minimum stimulator output required to elicit an MEP >50 µV (or a visible motor twitch) in at least five out of ten trials and as a %MSO [15], [16], [17].

B. MRI Collection and Head Model Creation
Participants underwent a high-resolution structural MRI (T1 MPRAGE sequence; 5 min) of a 1 mm section with no inter-slice gap as part of an expanded MRIguided TMS experiment obtained on a 3 tesla MRI scanner (MAGNETOM Skyra, Siemens) at NIMHANS, Bengaluru, India [Fig.1(a)].T1-weighted imaging was processed through SimNIBS (SimNIBS Developers 2019, v2.0.1) mri2mesh pipeline to create segmented head models.The resulting segments included gray matter [Fig.1(b)], white matter, skin, skull, cerebrospinal fluid, and ventricles.Segments were imported into Autodesk Meshmixer (AutoDesk Inc., v11.2.37),where abnormalities were smoothed [21].CCD was taken between the gray matter and the center of the coil at the PreCG.Total GMV was also calculated in Meshmixer [24].

C. Finite Element Analysis
Segmented head models were imported into Sim4Life, (Zurich Med Tech, v6.2.2) and were assigned individual material properties based on the IT'IS LF database (IT'IS Foundation, v4.1).An 11 turn figure-of-eight coil was modeled to match the Magventure MCF-B70 butterfly coil [Fig.2(a) and (b)] [31].Finite element analysis (FEA) was performed to match the empirical setup with the coil set at 3571 Hz.The coil was centered above the PreCG and placed 5 mm above the scalp to account for coil housing [24].Simulations were performed both at the maximum coil current, 5000 A and at each participant's RMT current.Simulations were performed with uniform grid spacing to ensure voxel size uniformity throughout the segments.Peak-induced EFS on the gray matter was visualized using the Sim4life surface viewer [Fig.2(c)] to confirm the maximum intensity was located on the PreCG.Once confirmed, the maximum EFS value on the gray matter was taken for both sets of simulations in V/m.DGMV was calculated as the percentage of gray matter voxels over the 100 V/m threshold out of all gray matter voxels and was also calculated in MATLAB (Mathworks, R2022a) [24].

D. Statistical Analysis
Two sample t-tests were performed in R (The R Foundation, v4.2.1) using the t-test function from the ggpubr library.t-tests were used to identify statistically significant differences between the healthy and schizophrenia populations.Linear regression modeling was also performed in R (The R Foundation, v4.2.1).Linear regression was performed between RMT Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.

III. RESULTS
We published data from this work on the Open Science Framework and is available at: https://osf.io/mka4f/?view_only=3cf42fffe4834e4c83f6840ce66a3f46 [32].

IV. DISCUSSION
The aim of this preliminary study was to investigate the role of participant-specific attributes and MRI-driven head Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.model parameters in accounting for RMT variability between schizophrenia and healthy participants.
Previous studies [33], [34] have reported values of 30-120 V/m as the amount of induced electric field required to initiate depolarization of the PreCG.In this study, we observed EFS-RMT to be 49.52 V/m (± 6.6 V/m) in simulations run to match the empirical coil strength for the schizophrenia population and 50.42 V/m (± 4.8 V/m) for the healthy population.This supports the notion of a constant depolarization threshold regardless of the participants' psychiatric condition.
The relationships of CCD with RMT, EFS-Max, and DGMV have been previously shown and were expected with linear regression modeling for both populations [24].The induced EFS decays rapidly as the distance between the TMS coil and the cerebral cortex increases [35].This supports the negative correlation found in this study of CCD with both DGMV and EFS.This also supports the positive correlation of CCD with RMT since the amount of current in the TMS coil needed to achieve depolarization also increases with distance.
The negative relationships of both EFS and DGMV at maximum current strength with RMT were expected within both populations.As EFS and DGMV increase, a smaller portion of the MSO is needed to induce the EFS required to depolarize neurons.
The hypothesis that schizophrenia would have a negative influence on the relationship between RMT and participantspecific attributes was supported.We found that the relationships of CCD, EFS, and DGMV with RMT were negatively impacted in the schizophrenia population versus the healthy population.Previous investigations have shown structural and functional alterations in the primary motor cortex due to the presence of schizophrenia [36].This suggests a need for the characterization of the role functional and structural connectivity influences TMS outcomes and RMT.
The lack of a statistical difference between the RMTs of the healthy and schizophrenia populations was unexpected.The schizophrenia population was expected to have an increased RMT over the healthy population.A potential explanation of this is the role of antipsychotic medications, which have shown influence on functional connectivity of the motor cortex in those with schizophrenia [37].The present study consisted of participants who have not previously used antipsychotic medications and have not been exposed to their effects on the motor cortex.
The negative influence of both EFS-Max and DGMV on PANSS was expected with linear regression modeling.As schizophrenia symptom severity increased, the amount of induced EFS and DGMV was expected to decrease when coil strength remained constant (5000 A).However, these relationships were expected to show a positive relationship between RMT and PANSS which we did not observe.Further investigation is needed to verify the direction, extent, and cause of these relationships.
This study was limited by the lack of T2-weighted imaging, diffusion tensor imaging (DTI), and functional connectivity.The inclusion of T2-weighted imaging would improve overall head model quality and resolution.DTI would allow for the creation of white matter-derived fiber tracts which have previously been shown to account for some of the variability of RMT [10].Alterations in white matter connections have been shown specifically in the motor regions of schizophrenia patients [36].Functional connectivity has been previously shown to influence RMT [23] and decrease in the motor region of those with schizophrenia [38].

V. CONCLUSION
The present study investigated how the primary TMS dosage parameter (RMT) is influenced by MRI model-driven parameters in 62 schizophrenia participants and 41 healthy participants.Our results show that empirically measured RMT is influenced by participant-specific neuroanatomical features in both healthy and schizophrenia participants.Our results also suggest that these relationships are weakened by the presence of schizophrenia.Further investigations are needed to determine the extent schizophrenia, functional connectivity, and white matter-derived fiber tractography influence these relationships.

Fig. 1 .
Fig. 1.(a) MRI scan of healthy participant 15 shown in the axial slice.(b) Meshmixer view of the segmented gray matter of healthy participant 15.

Fig. 2 .
Fig. 2. Sim4life viewer of healthy participant 15.(a) Segmented gray matter imported with coil positioning in coronal view.(b) Segmented gray matter imported into Sim4life with coil positioning over PreCG in axial view.(c) Sim4life surface viewer of the electric field distribution on gray matter.(d) Sim4life surface viewer of the magnetic field distribution on gray matter.

Fig. 3 .
Fig. 3. Coil-to-cortex correlated with RMT as a %MSO in the schizophrenia population with a regression coefficient R 2 = 0.429.

Fig. 6 .Fig. 7 .
Fig. 6.Coil to cortex correlated with RMT as a %MSO in the healthy population with a regression coefficient R 2 = 0.682.
March 2023; revised 12 May 2023; accepted 17 May 2023.Date of publication 5 June 2023; date of current version 24 October 2023.Corresponding author: R. L. Hadimani (e-mail: rhadimani@vcu.edu).This work involved human subjects or animals in its research.Approval of all ethical and experimental procedures and protocols was granted by the National Institute of Mental Health and Neuro-Sciences (NIMHANS) Human Ethics Committee under Approval No. NIMHANS/88TH IEC/2014.Color versions of one or more figures in this article are available at https://doi.org/10.1109/TMAG.2023.3282784.