Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/pinpointing-the-precise-stimulation-targets-for-brain-rehabilitation-gYlz4PuRZ0
- Publisher
- Springer Journals
- Copyright
- Copyright © The Author(s) 2023
- eISSN
- 1471-2202
- DOI
- 10.1186/s12868-023-00791-7
- Publisher site
- See Article on Publisher Site
Abstract
Background Transcranial magnetic stimulation ( TMS) is increasingly used as a promising non-pharmacological treatment for Parkinson’s disease (PD). Scalp-to-cortex distance (SCD), as a key technical parameter of TMS, plays a critical role in determining the locations of treatment targets and corresponding dosage. Due to the discrepancies in TMS protocols, the optimal targets and head models have yet to be established in PD patients. Objective To investigate the SCDs of the most popular used targets in left dorsolateral prefrontal cortex (DLPFC) and quantify its impact on the TMS-induced electric fields (E-fields) in early-stage PD patients. Methods Structural magnetic resonance imaging scans from PD patients (n = 47) and normal controls (n = 36) were drawn from the NEUROCON and Tao Wu datasets. SCD of left DLPFC was measured by Euclidean Distance in TMS Navigation system. The intensity and focality of SCD-dependent E-fields were examined and quantified using Finite Element Method. Results Early-stage PD patients showed an increased SCDs, higher variances in the SCDs and SCD-dependent E-fields across the seven targets of left DLPFC than normal controls. The stimulation targets located on gyral crown had more focal and homogeneous E-fields. The SCD of left DLPFC had a better performance in differentiating early-stage PD patients than global cognition and other brain measures. Conclusion SCD and SCD-dependent E-fields could determine the optimal TMS treatment targets and may also be used as a novel marker to differentiate early-stage PD patients. Our findings have important implications for developing optimal TMS protocols and personalized dosimetry in real-world clinical practice. Keywords Parkinson’s disease, Transcranial magnetic stimulation, Scalp-to-cortex distance, DLPFC, Head model, Simulation Data used in the preparation of this paper was obtained from NEUROCON and Tao Wu datasets (http://fcon_1000.projects.nitrc. org/indi/retro/parkinsons.html). The researchers at NEUROCON and Tao Wu datasets contributed to the design and implementation of NEUROCON and Tao Wu provided data but did not participate in writing of this paper. *Correspondence: Hanna Lu hannalu@cuhk.edu.hk Full list of author information is available at the end of the article © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Lu et al. BMC Neuroscience (2023) 24:24 Page 2 of 12 Introduction patients; (2) Intra-individual: differences in treatment The rapid ageing of populations around the world poses targets locations and corresponding variations in the an unprecedented set of challenges, the most significant SCDs of these targets. of which is the rise in the prevalence of age-related neu- Collectively, the longstanding absence of optimal treat- rodegenerative diseases, including Alzheimer’s disease ment targets has resulted in larger differences between (AD) and Parkinson’s disease (PD) [1]. Notably, the inci- clinicians regarding the TMS used in the treatment of dence rate of PD is increasing more quickly than that PD patients. With respect to the highly folded anatomical of AD [2]. The Global Burden of Disease Study predicts target (i.e., DLPFC), a rigorous quantitative approach to that the number of PD patients will double from around morphometric analysis is required to better quantify the 7 million in 2015 to about 13 million in 2040 [3]. Aside cortical features in the context of brain atrophy at indi- from the growing number of patients, PD is a compli- vidual level. Thus, the main aim of current study was to cated, progressive neurodegenerative disease with a wide systematically investigate the SCDs of the most popular spectrum of premotor, motor and non-motor symptoms used targets of left DLPFC and determine the optimal at different stages of the disease, but there are still very treatment targets in early-stage PD patients and age- few evidence-based therapeutic methods for managing matched normal controls. A second aim was to examine the depressive and cognitive symptoms in early-stage PD and quantify the SCD-dependent matrices of the TMS- patients [1]. Even though pharmacological therapies have induced E-fields using Finite Element Method (FEM). much helped to manage the motor symptoms [4], the negative consequences from dopamine administration to Materials and methods brain regions outside of the basal ganglia might cause or Participants exacerbate cognitive problems in PD patients [5, 6]. T1-weighted structural magnetic resonance imaging Based on well-established PD models, dopamine pro- (MRI) scans were aggregated from two publicly accessi- jections are predominantly directed to basal ganglia and ble datasets [16]: (1) The NEUROCON study included 27 prefrontal cortex (PFC) [7], therefore, the neuroanatomi- PD patients and 16 age-matched normal controls with- cal targets for brain stimulation are determined by the out a history of psychiatric or neurological diseases. (2) computational models of brain features. Transcranial The Tao Wu dataset included 20 PD patients and 20 age- magnetic stimulation (TMS), a non-invasive modality matched normal controls. According to the clinical stan- of brain stimulation in addition to deep brain stimula- dards of the Queen Square Brain Bank (QSBB) and the tion (DBS), has gained significant interests as a safe and European Federation of Neurological Societies / Move- effective treatment for the main types of age-related ment Disorder Society - European Section (EFNS/MDS- neurodegenerative diseases, including AD and PD [8, 9]. ES), all patients were in the early or moderate stage of PD The administration of TMS treatment over left dorsolat - (Hoehn and Yahr stages 1 to 2.5). We recruited 83 par- eral prefrontal cortex (DLPFC) has shown encouraging ticipants in total, comprising 47 early-stage PD patients results to enhance the cognition and motor functions and 36 age-matched normal controls. in PD patients [10–12]. However, these studies have In accordance with the ethical standards of the 1964 revealed a variety of findings [ 13]. The long-standing Declaration of Helsinki and its later amendments, the concerns regarding the heterogeneity derived from the NEUROCON study has been approved by the Univer- clinical trials continue to be fiercely discussed recently sity Emergency Hospital Bucharest ethics committee. [14]. The existing computational models of TMS are Each patient signed a written informed consent form to developed using the template from a general population, participate in the study. Additionally, the study sites spe- which may not closely resemble the head size and brain cifically secured consent for public sharing of the ano - features in senior adults or PD patients. Moreover, scalp- nymized data. The demographics of the participants, to-cortex distance (SCD), as a key technical parameter in terms of age, sex, and the years of education, and the of TMS, demonstrating the geometric distance from the scores of cognitive functions were directly obtained from scalp (i.e., TMS coil) to the cortex (i.e., cortical surface), the NEUROCON and Tao Wu datasets. The current has a significant detrimental impact on the electric fields study was approved by the Clinical Research Ethics Com- (E-fields) induced by TMS [ 15]. While developing opti- mittee of The Chinese University of Hong Kong (CUHK) mal TMS protocols is certainly important and therapeuti- and New Territories East Cluster (NTEC) (The Joint cally significant, the major challenges associated with the CUHK-NTEC). potential heterogeneity impede the personalized TMS treatment for the people living with early stage PD. The MRI acquisition heterogeneity caused by brain features may be decom- Details about the protocol of MRI acquisition can be posed into two levels: (1) Inter-individual: differences reviewed on the official webpage of Parkinson’s Disease in brain size and cortical features among early-stage PD Datasets (http://fcon_1000.projects.nitrc.org/indi/retro/ Lu et al. BMC Neuroscience (2023) 24:24 Page 3 of 12 parkinsons.html). The structural MRI scans derived from Surface-based morphometry analysis the NEUROCON study were acquired on a 1.5T Sie- Surface-based analysis of brain features, including cor- mens Avanto scanner with a thermo-plastic face mask tical volume, surface area and cortical thickness, were to minimize head movements [16]. To better co-regis- performed by BrainSuite 19a (http://brainsuite.org/) ter to standard Montreal Neurological Institute (MNI) (Fig. 1A). BrainSuite is an automatic cortical surface space, T1-weighted MRI scans were obtained for all identification integrated package with the refined ver - participants using a magnetization-prepared rapid gra- sion of brain surface extraction (BSE), which is suitable dient-echo (MPRAGE) sequence (IR method, TR = 1940 for individuals with brain atrophy [19, 20]. To extract and ms, TE = 3.08 ms, inversion time (IT) = 1100 ms, voxel quantify the region-specific morphometric features, we size 0.97 × 0.97 × 1 mm). The structural MRI scans from followed the standard BrainSuite pipeline with default the Tao Wu database was acquired on a 3.0T Siemens parameters and employed the parcellation scheme on Magnetom scanner. MPRAGE scans were obtained the basis of the Automated Anatomical Labeling (AAL) (TR = 1100 ms, TE = 3.39 ms, voxel size 1 × 1 × 1 mm) for template. the registration to the MNI space. Cortical thickness is calculated as an average of the dis- tance from the white matter (WM) surface to the closest Clinical and motor assessments point on the pial surface and from that point back to the The modified Hoehn and Yahr (HY) Scale was used to closest point to the WM surface. The measure of surface evaluate the clinical and motor symptoms and disease area is calculated using the triangular tessellation of the progression in PD patients. The HY scale was originally gray matter (GM) / WM interface (inner surface) and the described in 1967 and included five stages to PD. It has WM / Cerebrospinal fluid (CSF) interface (pial surface) since been modified with the addition of stages 1.5 and [21]. Cortical folding measured by gyrification index (GI) 2.5 to account for the intermediate course of PD [17]. is a ratio of inner surface area to the area of an outer sur- Mini-Mental State Examination (MMSE) was used to face that smoothly encloses the cortex [22]. evaluate the global cognitive function [18]. Fig. 1 The framework of MRI-based brain morphometric analysis and computational model of SCD-dependent TMS-induced electric fields. (A) Cortex- based features. After cortex reconstruction and segmentation, the quantitative cortical measures of treatment target derived from BrainSuite contained cortical thickness and surface area. (B) Scalp-based features. After constructing the scalp and cortex, we localized the seven off-site treatment targets and their locations on cortical surface, measured the scalp-to-cortex distance (SCD) and constructed the head models of SCD-dependent electric fields using Finite Element Method (FEM). Lu et al. BMC Neuroscience (2023) 24:24 Page 4 of 12 SCD of left DLPFC y = 39, z = 43; (2) Electroencephalography (EEG) F3: The MNI coordinates of the seven targets of left DLPFC x = − 37, y = 26, z = 49; (3) Average 5 cm: x = − 41, y = 16, were drawn from published studies (Fig. 2) [23–25], z = 54; (4) Fitzgerald Target: x = − 46, y = 45, z = 38; (5) including: (1) Brodmann Area (BA) 9 centre: x = − 36, Paus Cho Target: x = − 40, y = 31, z = 34; (6) Rusjan Target: Fig. 2 Comparisons of the scalp-to-cortex distance (SCD) in left dorsolateral prefrontal cortex (DLPFC) in early-stage PD patients and age-matched normal controls (NC). Data are displayed as mean ± SD. Seven targets of left DLPFC were located and marked on the scalp (A) and cortex (B), including: Brodmann Area (BA) 9 centre ( Target 1); EEG F3 ( Target 2); Average 5 cm ( Target 3); Fitzgerald Target ( Target 4); Paus Cho Target ( Target 5); Rusjan Target (Target 6); BA46 centre (Target 7). (C) The early-stage PD patients showed increased SCDs than normal controls across the seven targets of left DLPFC. Significant between-group differences of SCDs were found in Target 4 ( P = 0.036), Target 5 (P = 0.018) and Target 7 (P = 0.029) Lu et al. BMC Neuroscience (2023) 24:24 Page 5 of 12 Fig. 3 Illustrations of the target-specific scalp-to-cortex distance (SCD) and its variances in early-stage Parkinson’s disease (PD) patients and age-matched normal controls (NC). The spatial distributions of the seven treatment targets were demonstrated on the cortical area of left dorsolateral prefrontal cortex (DLPFC) in NC (A) and PD patients (B). Variances in the SCDs of the treatment targets of left DLPFC in PD patients (D) were higher than the SCDs in NC (C) (Grey dotted line represents the mean value of SCD across the seven targets). Data are displayed as mean ± SD. x = − 50, y = 30, z = 36; (7) BA46 centre: x = − 44, y = 40, between the coordinates locating on the scalp (x , y , z ) s s s z = 29. and the cortex (x , y , z ) in the MNI space with the fol- c c c Scalp-to-cortex distance (SCD), as a geometric feature, lowing formula [15, 27]: was measured in the Brainsight TMS neuronavigation system (http://thebrainx.com/landscape/) [26]. Based 2 2 2 ( ) ( ) ( ) D = x − x + y − y + z − z i s c s c s c on the structural MRI scans, we first reconstructed the 3D curvilinear of scalp and cortex, and then adjusted the MRI-to-head co-registration using the anterior commis- sure - posterior commissure (AC-PC) line in the MNI FEM model of SCD-dependent E-fields space. After co-registration, the locations of the targets To establish a realistic head model of TMS, we applied on cortex were identified and pinpointed with the MNI the Finite Element Method (FEM) [28, 29], a well- coordinates of the seven targets of left DLPFC (x, y, z). established approach for integrating different brain tis - To better mimic the realistic TMS treatment, the cor- sues and cortical surface features, thereby allowing it to responding locations of the seven targets of left DLPFC account for the impact of PD-related cortical changes. on the scalp were targeted in the neuronavigation system SimNIBS, as a state-of-the-art platform for the simula- by pointing the cursor to the scalp and then adjusting tion of transcranial brain stimulation (http://simnibs. the orientation of the TMS coil from the midline at 45°. de/), allows for the computational calculations of the Euclidean distance (D ) was used to measure the distance TMS-induced E-fields [ 30]. As part of SimNIBS pipeline, i Lu et al. BMC Neuroscience (2023) 24:24 Page 6 of 12 FEM model distinguishes between scalp, skull, CSF, GM corrections of multiple comparisons were performed by and WM, of which the assigned values of conductivities the above code using false discovery rate (FDR) estima- are σ = 0.465 S/m, σ = 0.01 S/m, σ = 1.654 S/m, tion; 2-sided p < 0.05 is considered statistically significant. skin skull CSF σ = 0.276 S/m, and σ = 0.126 S/m [31]. The quantitative measures of SCD-dependent E-fields GM WM The first step in constructing the realistic head model were compared between the seven targets of left DLPFC was to generate a conductor model of the head. In order using repeated measures analysis of variance (ANOVA), to create the finite element mesh, we assigned each voxel followed by post-hoc t-tests with Bonferroni correction. in structural MRI scans to a specific tissue type (Fig. 1B). The receiving operating characteristic (ROC) analysis As a recommended option in SimNIBS, we selected was conducted to evaluate the power of cognitive and headreco in combination with the Statistical Parametric brain measures in differentiating the individuals with dif - Mapping (SPM12) toolbox (https://www.fil.ion.ucl.ac.uk/ ferent clinical statuses. The χ test, ANOVA, Pearson cor- spm/software/spm12/) for achieving an accurate seg- relation coefficients and ROC analysis were performed mentations of brain tissues. The second step was to clean using R (version 3.6.2) and R Studio (version 1.3.959) the tissue maps by applying morphological operations, software. and then use the tissue maps to create surface recon- structions. Finally, the FEM mesh was generated by filling Results in the tetrahedrons between the surfaces of tissue using Demographics, cognitive function, and brain Gmsh (http://gmsh.info/). morphometry After constructing the realistic head model, the simula- Demographics in terms of age and sex, global cognition tion of single-pulse TMS begin by adding the coil (Mags- (measured by MMSE), global brain morphometry and tim 70 mm figure-8 coil) to the scalp. In this step, targets the morphometric features of left DLPFC, including GM on the scalp were shifted to form the shape of the coil, volume, WM volume, surface area, cortical thickness and while keeping good quality elements. Afterwards, the folding (measured by gyrification index), were compara - body of the coil was constructed by filling in tetrahedra. ble between PD patients and normal controls (Table 1). Simulations were run with a TMS pulse of 1.00 × 10e6 A/s, a Magstim 70 mm figure-8 coil over the targets of Comparisons of the SCDs in left DLPFC left DLPFC (i.e., F3 in International 10–20 system but Compared to normal controls, PD patients had increased modified with individual MNI coordinates on the cortex). SCDs across the seven targets of left DLPFC (Table 2). Default conductivities of the toolbox were used for the As depicted in Fig. 2B, significant differences of SCDs different compartments as mentioned above [ 31]. were found in specific targets, including Fitzgerald Tar - get (Target 4: t = -2.14, p = 0.036), Paus Cho Target (Tar- Quantitative measures of SCD-dependent E-fields get 5: t = -2.42, p = 0.018), and BA 46 centre (Target 7: t Considering the E-fields refer to a vector field, both = -2.23, p = 0.029). Within each group, the SCD of Paus intensity and focality of the simulated E-fields are quanti - Cho Target was significantly greater than the other tar - fied and visualized as norm or strength (i.e., vector length gets (Normal controls: p = 0.005; PD patients: p < 0.001). or magnitude) [31]. The E-fields magnitude of each tar - To better understand the heterogeneity in SCD, we fur- get was quantified as the 95%, 99% and 99.9% of E-fields ther calculated the variance of the SCDs of the seven tar- (x −x ) mean target strength (Norm E) (Fig. 4A). To avoid the outlier effects, gets of left DLPFC using the formula: S = N−1 the peak value of E-fields magnitude ( E ) is defined as , x represents the SCD of each target, x repre- max target mean the 99.9% of E-fields strength. The focality of E-fields is sents the average value of SCD across the seven targets, measured as the GM volume with the E-fields greater or N represents the number of targets. Compared to normal equal to 50-75% of the peak value. TMS-affected cortical controls, PD patients showed higher variability in SCDs volume was quantified as the volume corresponding to between the seven targets of left DLPFC (S : Normal con- the 50% (Foc ) and 75% (Foc ) of the maximum E-fields. trols: 2.266 ± 1.623, PD patients: 3.472 ± 2.458, t = -2.152, 50 75 p = 0.034) (Fig. 3). Statistical analysis The differences of demographics, clinical features, cog - Quantitative measures of SCD-dependent E-fields nitive performance and the SCDs of left DLPFC were SCD-dependent E-fields intensity and focality were simu - tested either with the chi-square (χ ) test for categori- lated and quantified for the seven targets of left DLPFC, cal variable or with independent two samples t-test for indicating a heterogeneous pattern of E-fields distri - continuous variables. The groupwise comparisons of bution across the targets (Supplementary Fig. 1). The the morphometric features were conducted by using the average E in PD patients was 0.858 V/m, which was max code embedded in the MATLAB (http://neuroimage. weaker than the value in normal controls (0.908 V/m). usc.edu/neuro/Resources/BST_SVReg_Utilities). The Significant differences of the magnitude and current Lu et al. BMC Neuroscience (2023) 24:24 Page 7 of 12 Fig. 4 Comparisons of the magnitude and current intensity of the SCD-dependent TMS-induced E-fields in normal controls (NC) and Parkinson’s disease (PD) patients. Data are displayed as mean ± SD. (A) The magnitude of the E-fields was measured as the 95% (yellow), 99% (orange) and 99.9% ( E ) (red) max of the strength of simulated E-fields. The early-stage PD patients showed significant decreased E (V/m) (NormE) (B) and current intensity (NormJ) max (A/m ) (C) in Target 3 (EEG F3), Target 6 (Rusjan Target) and Target 7 (BA46 centre). SCD, Scalp-to-cortex distance; TMS, Transcranial magnetic stimulation; E-fields, Electric fields density of the SCD-dependent E-fields were found in PD patients and 5.53 cm in normal controls. The Foc of specific targets, including EEG F3 (Target 2: t = 2.129, the SCD-dependent E-fields varied between 5.16 cm and p = 0.041), Average 5 cm (Target 3: t = 2.919, p = 0.006), 6.31 cm (Supplementary Table 3), which corresponded Rusjan Target (Target 6: t = 2.936, p = 0.006), BA 46 cen- to 38.5-46.5% of the GM volume in left DLPFC in normal tre (Target 7: t = 2.298, p = 0.028) (Fig. 4) (Supplementary controls and 39.4-47.8% of the GM volume in left DLPFC Tables 1 and Table 2). The average Foc was 5.59 cm in in PD patients. Among the seven targets of left DLPFC, 75 Lu et al. BMC Neuroscience (2023) 24:24 Page 8 of 12 Table 1 Demographics, clinical symptoms and global p = 0.769), indicating that PD patients had heterogeneous morphometric features distribution of the E-fields in TMS-affected cortical vol - Normal PD t p ume (Supplementary Fig. 2). controls patients (χ ) value (n = 36) (n = 47) Associations between SCD and clinical features Age (years) 66.03 ± 8.92 67.21 ± 8.61 -0.612 0.542 Using age and sex as covariates, the relationship between Sex (M/F) 16:20 19:28 0.895 0.611 SCD, the magnitude and focality of the E-fields and MMSE 29.05 ± 1.32 28.75 ± 1.07 0.791 0.434 morphometric features was examined at each target of HY - 1.91 ± 0.47 - - left DLPFC. We found that the gyrification index of left Global brain morphometry DLPFC was significantly correlated with the target-spe - Mean CT (mm) 3.95 ± 0.21 3.98 ± 0.24 -0.751 0.455 cific SCD in PD patients, including the Paus Cho Target 3 3 Mean GMV (×10 mm ) 6.49 ± 0.59 6.55 ± 0.68 -0.397 0.692 (Target 5: r = 0.384, p = 0.008), Rusjan Target (Target 6: 3 3 Mean WMV (×10 mm ) 3.83 ± 0.54 3.91 ± 0.59 -0.591 0.556 r = 0.348, p = 0.017) and BA 46 centre (Target 7: r = 0.356, Morphometry of left p = 0.014), but not in normal controls. DLPFC CT (mm) 4.39 ± 0.42 4.38 ± 0.41 0.067 0.947 ROC analysis 3 3 GMV (×10 mm ) 13.39 ± 1.75 13.27 ± 1.99 0.301 0.764 To classify the individuals with PD, the value of the area 3 3 WMV (×10 mm ) 6.86 ± 1.37 7.05 ± 1.67 -0.564 0.574 under the ROC curve (AUC) was used to test the dis- Pial surface area (×10 5.53 ± 0.69 5.49 ± 0.94 0.261 0.796 criminant power of the brain and clinical features. We mm ) found that neither MMSE nor other brain measures Inner surface area 3.28 ± 0.61 3.27 ± 0.68 0.092 0.927 3 2 showed a significant discriminative power (Supplemen - (×10 mm ) tary Fig. 3); while the geometric measure of the mean Gyrification index 1.71 ± 0.22 1.70 ± 0.21 0.252 0.801 SCD of left DLPFC, had a better performance to differ - Note. Data are raw scores and presented as mean ± SD. entiate PD patients from normal controls (AUC = 0.733, Abbreviations: PD = Parkinson’s disease; MMSE = The Mini-Mental State Exam; HY = Hoehn and Yahr Scale; CT = Cortical thickness; GMV = Gray matter volume; p = 0.012). Moreover, all the SCDs of left DLPFC targets WMV = White matter volume showed significant power to discriminate PD patients. Table 2 Scalp-to-cortex distance (SCD) of the seven targets of Discussion left DLPFC TMS is a non-invasive brain stimulation technology that SCD Normal PD t p is being increasingly employed as a non-pharmacological controls patients (χ ) value treatment for age-related neurodegenerative diseases. (n = 36) (n = 47) The variety of treatment targets and the concomitant 1. BA9 center 14.75 ± 2.68 15.71 ± 2.26 -1.78 0.079 E-fields-dependent dosimetry limits the personalized 2. EEG F3 14.53 ± 2.67 15.53 ± 2.32 -1.83 0.072 applications of TMS in clinical practice. In this study, 3. Average 5 cm 14.24 ± 2.64 15.08 ± 2.31 -1.55 0.125 we examined and quantified the scalp-to-cortex distance 4. Fitzgerald Target 14.04 ± 2.49 15.11 ± 2.02 -2.14 0.036 (SCD) of seven commonly used targets of left DLPFC and 5. Paus Cho Target 16.28 ± 3.98 18.02 ± 2.52 -2.42 0.018 its impact on the intensity and focality of the simulated 6. Rusjan Target 14.03 ± 2.49 14.94 ± 2.02 -1.85 0.069 7. BA 46 center 14.06 ± 2.76 15.22 ± 1.99 -2.23 0.029 E-fields through computational realistic head models in Note. Data are raw scores and presented as mean ± SD. early-stage PD patients. Abbreviations: DLPFC = Dorsolateral Prefrontal Cortex; PD = Parkinson’s Disease; BA = Brodmann Area; EEG = Elec troencephalography. Heterogeneity in treatment targets With the rapid advances in imaging and analytical tools, the peak values were significantly decreased in Rusjan the stimulation-targeting rules have been switched from Target (Target 6: t = -3.151, p = 0.003) and BA 46 center scalp-based to cortex / laminar-specific targeting [ 32, (Target 7: t = -2.319, p = 0.027) in PD patients. 33]. In recent years, non-invasive neuroimaging tech- As to the heterogeneity in TMS-induced E-fields focal - nologies, such as MRI, functional near-infrared spectros- ity, we calculated the variance of the SCD-dependent copy (fNIRS) and positron emission tomography (PET), (x −x ) mean target E-fields using the same formula: , offer great promise for determining the TMS treatment S = N−1 x represents the target-specific E-fields, x repre- targets at individual level. Using functional MRI as an target mean sents the average value of E-fields across the seven tar - example, functional connectivity-based DLPFC target- gets, N represents the number of targets. Compared to ing for the treatment of major depressive disorders has normal controls, PD patients showed comparable vari- been well developed and evaluated [23, 34, 35]. Although ance in the Foc of the SCD-dependent E-fields (Normal there were significant differences in target-based func - controls: 0.2 ± 0.03, PD patients: 0.21 ± 0.02, t = -0.297, tional connectivity among DLPFC targets [25, 36, 37], the Lu et al. BMC Neuroscience (2023) 24:24 Page 9 of 12 heterogeneity in targeting highlights the necessity and upon neurodegeneration and may have potential clinical importance of pinpointing the optimal stimulation sites utilities. within the DLPFC rather than a general anatomical area. Despite these informative findings, a pertinent question SCD as new marker for PD patients is concerned with the more efficient strategy for deter - At individual level, although the predefined MNI coor - mining the precise targets for TMS treatment or, alterna- dinates and the labeled SCD-adjusted targets on the cor- tively, whether optimal targeting within a TMS target is a tex were located within the surface of left DLPFC, the quick and plausible way for clinical populations, particu- variations in the SCDs across the targets indicate greater larly those with neurodegenerative diseases. intra-individual variability in geometric features and the Based on the published TMS studies in PD patients, a corresponding E-fields focality in PD patients. At the figure-8 coil was positioned over the left DLPFC through group level, increased SCDs, rather than global cognition scalp-based targeting, including Average 5 cm [38, 39], and morphometric features, demonstrated strong posi- EEG F3 [11], or neuronavigated targeting with the MNI tive correlations with the cortical folding of treatment coordinates close to BA46 centre [10, 12]. However, when target, as well as a significant power to differentiate early- comes to senior adults or patients with dementia, there stage PD patients from age- / morphometry-matched are several issues that need to be addressed in consider- normal controls, implying that the vector-like SCD, as a ing the TMS treatment targets. First, it is critical to spe- key technical parameter, is more than simply useful for cifically map the cortical site for stimulation: whether guiding TMS coil (i.e., scalp site) to the precise target this is a set of predefined MNI coordinates within the located on the highly folded cortex. Of note, the impacts DLPFC, or a specific surface on the cortex. Second to this of morphometric features on the simulated E-Fields, par- is the approach to ensure the TMS coil is positioned over ticularly cortical folding, have been tested in the studies the corresponding scalp site with right orientation over of TMS [48], and other modalities of transcranial brain the stimulation site of left DLPFC. Third, it is critical to stimulation [49]. Therefore, this intrinsically connects localize the cortical target in combination with the sur- geometric and morphometric features (i.e., SCD and cor- face features of DLPFC. tical folding) with precise identification of the stimula - The surface features of cortical targets vary from region tion targets, which makes it possible for optimizing the to region in the individuals with different cognitive sta - process of TMS targeting in combination with the under- tuses. Unlike AD and Frontotemporal dementia (FTD), lying surface morphometry. PD is not a typical age-related neurodegenerative dis- Our findings show that the targets close to the gyral ease with cortical atrophy [40, 41]. In cognitively normal crown of the left DLPFC, such as Average 5 cm, EEG early-stage PD patients, there was little or no cortical F3, BA9 centre, and Rusjan Target, had stronger E-fields atrophy or thinning as compared to normal controls. intensity than the other targets, which are consistent with Indeed, the results of global morphometry and the mor- previous findings [ 50, 51]. Meanwhile, we found that the phometric features of left DLPFC in PD patients accord SCDs of the inferior targets of left DLPFC, such as Paus with the documented imaging studies [41, 42]. Although Cho Target, Rusjan Target and BA 46 centre, are signifi - PD patients had comparable brain volumes and cortical cantly associated with the higher degree of cortical fold- thickness, the SCD of left DLPFC in PD patients were ing, but not other morphometric features. It should be generally greater than in normal controls. It should be noted that BA9 and BA46 are well-established DLPFC noted that the distance between the TMS coil (i.e., scalp) subregions with diverse cytoarchitecture and neural cir- to the cortex is a key technical parameter that determines cuits among the seven targets [52]. For instance, BA9 and the motor-evoked potential (MEP) and the output of BA46 are disproportionately affected in brain disorders, TMS (i.e., region-specific dosage) [ 43–45]. Furthermore, with BA46 showing the most predominant layer II thin- even though TMS coils vary in type and size, the range ning, but not BA9 [50]. In our results, the distinct pat- of TMS penetration depth is around 0.9–3.5 cm [46], fig - terns of SCD-dependent intensity and focality of the ure-8 coil has a penetration depth of 1.5 to 2.5 cm [47]. E-fields in early-stage PD patients highlighted the fea - Six of the seven targets within left DLPFC had SCDs sibility, necessity and importance of targeting the left greater than 1.5 cm. Of note, the Paus Cho Target (Target DLPFC subregions through realistic geometric model- 5) located at the sulcal crown with a greatest SCD had the ing for the first time. As a result, for transcranial brain lower intensity and focality of the TMS-induced E-fields stimulation, region-specific surface-based brain features than the other targets. Interestingly, in the context of provide valuable information of the cortical landscape global increased SCD, the heterogeneity in the SCDs underneath TMS coil, which is expected to combine of the targets within left DLPFC was also significantly morphometric and geometric features in the construc- increased in early-stage PD patients, which is thought to tion of personalized head model. be a preferentially affected neurophenotype depending Lu et al. BMC Neuroscience (2023) 24:24 Page 10 of 12 Our findings might pave the way for personalized tran - may influence the stimulation dose at individual level. For scranial brain stimulation that targets the disease-specific addressing this issue, future work will in-depth investi- optimal targets using geometric feature-informed head gate the geometric features of the treatment targets and models. At a minimum, our study supports a quick and plug them in the optimization of coil placement and plausible assessment of SCD for localizing the treatment parameter estimations in transcranial brain stimulation target sites on the scalp and cortex for senior adults and studies. Last but not least, the MRI-informed head model PD patients. This may enable disease-specific selection should be validated and used in other advanced modal- of stimulation sites and TMS dosage adjustment across ity of brain stimulation, such as transcranial ultrasound different brain regions. In prior research, we found stimulation (TUS) [53, 54]. region-specific SCD and its impacts on E-fields in nor - mal ageing people, as well as patients with mild cognitive Conclusion impairment (MCI) and AD [16, 27]. The current work In conclusion, the current work adds to the growing may potentially be useful in developing an advanced, knowledge that individual brain features can aid in pre- next-generation therapeutic framework of personalized cisely localizing treatment targets and optimizing the brain stimulation for patients with neurodegenerative dose for transcranial brain stimulation. An integrated diseases. The SCD and SCD-dependent E-fields could be imaging-informed computational model allows brain partitioned into a pre-treatment parameter for optimiz- measures and electric fields matrices to simultane - ing stimulation dose in computational E-fields modeling. ously inform one another, resulting in a more radiomic Based on the findings, we recommend that the treatment and realistic understanding of optimal TMS targeting in targets located on gyral crown with a shorter SCD have a senior adults and the individuals suffering from Parkin - higher intensity and focality of E-fields than the other tar - son’s disease. The region-specific brain features identified gets for modulating or enhancing motor function, cogni- here provide a new direction for dissecting the heteroge- tive functions, sleep quality and mood in early-stage PD neity, complementing previous attempts to develop per- patients. sonalized treatment strategies. Abbreviations Limitations and future directions AAL Automated Anatomical Labeling The current work illustrates the heterogeneity in region-AD Alzheimer’s disease BA Brodmann area specific SCD and the corresponding stimulation-induced DLPFC Dorsolateral prefrontal cortex electric fields intensity in early-stage PD patients, how - EEG Electroencephalography ever the findings should be interpreted with respect to E-fields Electric fields FEM Finite Element Method several limitations. First, this study examined the brain FTD Frontotemporal dementia features in early-stage PD patients and normal controls GM Gray matter from two separate datasets. Although the inclusion and HY Hoehn and Yahr Scale MCI Mild cognitive impairment exclusion criteria for PD patients were fairly standard, MEP Motor-evoked potential and the brain features of PD patients in two datasets MMSE Mini-mental state examination were comparable, the cognitive status might have influ - MNI Montreal Neurological Institute MPRAGE Magnetization-prepared rapid gradient-echo enced the results. Besides, because the sample number is MRI Magnetic resonance imaging moderately low, a bigger dataset for validation is neces- PD Parkinson’s disease sary before applying to clinical practice. Second, because PFC Prefrontal cortex SCD Scalp-to-cortex distance the current findings are based on cross-sectional data, ROC Receiving operating characteristic they cannot be extrapolated to ageing effects or disease TMS Transcranial magnetic stimulation progression. Furthermore, several key variables linked to TUS Transcranial ultrasound stimulation WM White matter metabolism and genetic risk factors were not provided in the NEUROCON and Tao Wu datasets, limiting the capacity to examine the genetic determinants of the brain Supplementary Information The online version contains supplementary material available at https://doi. (i.e., neurogenetics). org/10.1186/s12868-023-00791-7. Future research should validate the current results in a bigger PD dataset with genetic information, such as Additional file 1: Supplementary Table 1 . The magnitude and distribu- tion of SCD-dependent electric fields in PD patients and normal controls. Parkinson’s Progression Markers Initiative (PPMI) and Supplementary Table 2. The distribution of SCD-dependent electric examine the heterogeneity of geometric and morpho- fields in PD patients and normal controls. Supplementary Table 3. The fo- metric brain measures in different age groups and other cality of SCD-dependent electric fields in PD patients and normal controls. Supplementary Figure 1. Head models of transcranial magnetic stimula- main types of neurodegenerative diseases (e.g., AD, fron- tion ( TMS)-induced SCD-dependent electric fields (E-fields) in early-stage totemporal dementia). In terms of technique, it is critical PD patients. Supplementary Figure 2. Comparisons of the focality of the to note that the scalp-to-cortex distance and gyrification Lu et al. BMC Neuroscience (2023) 24:24 Page 11 of 12 SCD-dependent transcranial magnetic stimulation ( TMS)-induced electric National Laboratory of Pattern Recognition, Institute of Automation, fields (E-fields) in normal controls (NCs) and early-stage Parkinson’s disease Chinese Academy of Sciences, Beijing, China (PD) patients. Supplementary Figure 3. Receiver-operator characteristic Research Center for Augmented Intelligence, Zhejiang Lab, (ROC) curves for the cognition and the geometric and morphometric Hangzhou 311100, China features with differential values in early-stage Parkinson’s disease (PD) patients. Received: 29 August 2022 / Accepted: 15 March 2023 Acknowledgments The authors would like to thank the Principal Investigators of the NEUROCON and Tao Wu dataset, for providing clinical and MRI data. The NEUROCON was supported by NIH grants P50 AG05681, P01 AG03991, P01 AG026276, R01 References AG021910, P20 MH071616, U24 RR021382. The funders were not involved in 1. Jankovic J, Tan EK. Parkinson’s disease: etiopathogenesis and treatment. J the study design, data collection and analysis, or the preparation of the article. Neurol Neurosurg Psychiatry. 2020;91:795–808. 2. Dorsey ER, Bloem BR. The Parkinson Pandemic-A call to action. JAMA Neurol. Author contributions 2018;75:9–10. Lu initiated and organized the project. Li organized the dataset and references. 3. GBD 2016 Parkinson’s Disease Collaborators. Global, regional, and national Zhang, Meng, Ning and Jiang reviewed and revised the manuscript. All burden of Parkinson’s disease, 1990–2016: a systematic analysis for the global authors read and approved the final manuscript and have given final approval burden of Disease Study 2016. Lancet Neurol. 2018;17:939–53. of the current version. 4. Connolly BS, Lang AE. Pharmacological treatment of Parkinson disease: a review. JAMA. 2014;311:1670–83. Funding 5. Williams-Gray CH, Mason SL, Evans JR, Foltynie T, Brayne C, Robbins TW, The authors disclosed receipt of the following financial support for the Barker RA. The CamPaIGN study of Parkinson’s disease: 10-year outlook research: This work was supported by the Direct Grants for Research the in an incident population-based cohort. J Neurol Neurosurg Psychiatry. Chinese University of Hong Kong (Project Number: 4054569, 4054645) and the 2013;84:1258–64. Open Project Fund of the National Laboratory of Pattern Recognition, Chinese 6. Kim HJ, Mason S, Foltynie T, Winder-Rhodes S, Barker RA, Williams-Gray CH. Academy of Sciences (CAS) (Project Number: 202100034). The corresponding Motor complications in Parkinson’s disease: 13-year follow-up of the Cam- author had full access to all the data in the study and had final responsibility PaIGN cohort. Mov Disord. 2020;35:185–90. for the decision to submit for publication. 7. Maiti P, Manna J, Dunbar GL. Current understanding of the molecular mechanisms in Parkinson’s disease: targets for potential treatments. Transl Availability of data and materials Neurodegener. 2017;6:1–35. The datasets used and analyzed during the current study are available from 8. Weiler M, Stieger KC, Long JM, Rapp PR. Transcranial Magnetic Stimulation in the corresponding author on reasonable request. Alzheimer’s Disease: Are We Ready?eNeuro.2020;7. 9. Yuan TF, Li WG, Zhang C, Wei H, Sun S, Xu NJ, Liu J, Xu TL. Targeting neuroplas- Declarations ticity in patients with neurodegenerative diseases using brain stimulation techniques. Transl Neurodegener. 2020;9:1–10. Ethics approval and consent to participate 10. Trung J, Hanganu A, Jobert S, Degroot C, Mejia-Constain B, Kibreab M, Bru- For the aims of this study, we used the structural MRI data of the NEUROCON neau MA, Lafontaine AL, Strafella A, Monchi O. Transcranial magnetic stimula- and Tao Wu dataset that was previously collected under the study protocols tion improves cognition over time in Parkinson’s disease. Parkinsonism Relat at the Washington University. This study was carried out in accordance with Disord. 2019;66:3–8. the recommendations of the University’s Institutional Review Board (IRB). All 11. Hill AT, McModie S, Fung W, Hoy KE, Chung SW, Bertram KL. Impact of participants gave written informed consent at the time of study participation prefrontal intermittent theta-burst stimulation on working memory and in accordance with the Declaration of Helsinki. The University’s IRB approved executive function in Parkinson’s disease: a double-blind sham-controlled the research protocol and provided the explicit approval for the open sharing pilot study. Brain Res. 2020;1726:146506. of the anonymized data. The current study was approved by the Clinical 12. Lang S, Gan LS, Yoon EJ, Hanganu A, Kibreab M, Cheetham J, Hammer T, Research Ethics Committee of The Chinese University of Hong Kong (CUHK) Kathol I, Sarna J, Martino D, Monchi O. Theta-Burst Stimulation for Cognitive and New Territories East Cluster (NTEC) ( The Joint CUHK-NTEC). Enhancement in Parkinson’s Disease with mild cognitive impairment: a Randomized, Double-Blind, sham-controlled trial. Front Neurol. 2020;11:1461. Consent for publication 13. Li R, He Y, Qin W, Zhang Z, Su J, Guan Q, Jin L. Eec ff ts of Repetitive Transcranial Not applicable. Magnetic Stimulation on Motor Symptoms in Parkinson’s Disease: A Meta- Analysis. Neurorehabil Neural Repair. 2022;15459683221095034. Competing interests 14. Polanía R, Nitsche MA, Ruff CC. Studying and modifying brain function with The authors declared no potential conflicts of interest with respect to the non-invasive brain stimulation. Nat Neurosci. 2018;21:174–87. research, authorship, and /or publication of this article. 15. Lu H, Chan SSM, Lam LCW. Localized analysis of Normalized Distance from Scalp to cortex and personalized evaluation (LANDSCAPE): focusing on Age- Author details and dementia-specific changes. J Alzheimers Dis. 2019;67:1331–41. G27, Multi-Centre, Department of Psychiatry, The Chinese University of 16. Badea L, Onu M, Wu T, Roceanu A, Bajenaru O. Exploring the reproducibil- Hong Kong, Tai Po Hospital, Hong Kong SAR, China ity of functional connectivity alterations in Parkinson’s disease. PLoS ONE. Centre for Neuromodulation and Rehabilitation, The Affiliated Brain 2017;12:e0188196. Hospital of Guangzhou Medical University, Guangzhou, China 17. Goetz CG, Poewe W, Rascol O, Sampaio C, Stebbins GT, Counsell C, Giladi N, The Affiliated Brain Hospital of Guangzhou Medical University, Holloway RG, Moore CG, Wenning GK, Yahr MD, Seidl L. Movement Disorder Guangzhou, China Society Task Force report on the Hoehn and Yahr staging scale: status and Department of Mechanical and Automation Engineering, The Chinese recommendations. Mov Disord. 2004;19:1020–8. University of Hong Kong, Hong Kong SAR, China 18. Zadikoff C, Fox SH, Tang-Wai DF, Thomsen T, de Bie RM, Wadia P, Miyasaki J, Academy of Medical Engineering and Translational Medicine, Tianjin Duff-Canning S, Lang AE, Marras C. A comparison of the mini mental state University, Tianjin, China exam to the Montreal cognitive assessment in identifying cognitive deficits The First School of Clinical Medicine, Southern Medical University, in Parkinson’s disease. Mov Disord. 2008;23:297–9. Guangzhou, China 19. Shattuck DW, Mirza M, Adisetiyo V, Hojatkashani C, Salamon G, Narr KL, Brainnetome Center, Institute of Automation, Chinese Academy of Poldrack RA, Bilder RM, Toga AW. Construction of a 3D probabilistic atlas of Sciences, Beijing, China human cortical structures. NeuroImage. 2008;39:1064–80. Lu et al. BMC Neuroscience (2023) 24:24 Page 12 of 12 20. Lu H, Ma SL, Chan SS, Lam LC. The effects of apolipoprotein ε 4 on aging 39. Bentley JN, Irwin ZT, Black SD, Roach ML, Vaden RJ, Gonzalez CL, Khan AU, brain in cognitively normal chinese elderly: a surface-based morphometry El-Sayed GA, Knight RT, Guthrie BL, Walker HC. Subcortical intermittent Theta- study. Int Psychogeriatr. 2016;28:1503–11. Burst Stimulation (iTBS) increases Theta-Power in Dorsolateral Prefrontal 21. Cao B, Mwangi B, Passos IC, Wu MJ, Keser Z, Zunta-Soares GB, Xu D, Hasan Cortex (DLPFC). Front Neurosci. 2020;14:41. KM, Soares JC. Lifespan gyrification trajectories of human brain in healthy 40. Zarei M, Ibarretxe-Bilbao N, Compta Y, Hough M, Junque C, Bargallo N, Tolosa individuals and patients with Major Psychiatric Disorders. Sci Rep. 2017;7:1–8. E, Martí MJ. Cortical thinning is associated with disease stages and dementia 22. Madan CR, Kensinger EA. Cortical complexity as a measure of age-related in Parkinson’s disease. J Neurol Neurosurg Psychiatry. 2013;84:875–81. brain atrophy. NeuroImage. 2016;134:617–29. 41. Sterling NW, Lewis MM, Du G, Huang X. Structural imaging and Parkinson’s 23. Fox MD, Buckner RL, White MP, Greicius MD, Pascual-Leone A. Efficacy of Disease: moving toward quantitative markers of Disease Progression. J Parkin- transcranial magnetic stimulation targets for depression is related to intrinsic sons Dis. 2016;6:557–67. functional connectivity with the subgenual cingulate. Biol Psychiatry. 42. Laansma MA, Bright JK, Al-Bachari S, Anderson TJ, Ard T, Assogna F, Baquero 2012;72:595–603. KA, Berendse HW, Blair J, Cendes F, Dalrymple-Alford JC, de Bie RMA, Debove 24. Fitzgerald PB. Targeting repetitive transcranial magnetic stimulation in I, Dirkx MF, Druzgal J, Emsley HCA, Garraux G, Guimarães RP, Gutman BA, depression: do we really know what we are stimulating and how best to do Helmich RC, Klein JC, Mackay CE, McMillan CT, Melzer TR, Parkes LM, Piras F, it? Brain Stimul. 2021;14:730–6. Pitcher TL, Poston KL, Rango M, Ribeiro LF, Rocha CS, Rummel C, Santos LSR, 25. Cash RFH, Weigand A, Zalesky A, Siddiqi SH, Downar J, Fitzgerald PB, Fox MD. Schmidt R, Schwingenschuh P, Spalletta G, Squarcina L, van den Heuvel OA, Using brain imaging to improve spatial targeting of Transcranial magnetic Vriend C, Wang JJ, Weintraub D, Wiest R, Yasuda CL, Jahanshad N, Thompson stimulation for Depression. Biol Psychiatry. 2021;90:689–700. PM, van der Werf YD. International Multicenter Analysis of Brain structure 26. Lu H, Li J, Zhang L, Chan SSM, Lam LCW. Dynamic changes of region-specific across clinical stages of Parkinson’s Disease. Mov Disord. 2021;36:2583–94. cortical features and scalp-to-cortex distance: implications for transcranial 43. Nahas Z, Li X, Kozel FA, Mirzki D, Memon M, Miller K, Yamanaka K, Anderson B, current stimulation modeling. J Neuroeng Rehabil. 2021;18:1–12. Chae JH, Bohning DE, Mintzer J, George MS. Safety and benefits of distance- 27. Randazzo MJ, Kondylis ED, Alhourani A, Wozny TA, Lipski WJ, Crammond DJ, adjusted prefrontal transcranial magnetic stimulation in depressed patients Richardson RM. Three-dimensional localization of cortical electrodes in deep 55–75 years of age: a pilot study. Depress Anxiety. 2004;19:249–56. brain stimulation surgery from intraoperative fluoroscopy. NeuroImage. 44. Stokes MG, Chambers CD, Gould IC, Henderson TR, Janko NE, Allen NB, Mat- 2016;125:515–21. tingley JB. Simple metric for scaling motor threshold based on scalp-cortex 28. Ziegler E, Chellappa SL, Gaggioni G, Ly JQM, Vandewalle G, André E, Geuzaine distance: application to studies using transcranial magnetic stimulation. J C, Phillips C. A finite-element reciprocity solution for EEG forward modeling Neurophysiol. 2005;94:4520–7. with realistic individual head models. NeuroImage. 2014;103:542–51. 45. Stokes MG, Chambers CD, Gould IC, English T, McNaught E, McDonald O, 29. Gomez LJ, Dannhauer M, Koponen LM, Peterchev AV. Conditions for numeri- Mattingley JB. Distance-adjusted motor threshold for transcranial magnetic cally accurate TMS electric field simulation. Brain Stimul. 2020;13:157–66. stimulation. Clin Neurophysiol. 2007;118:1617–25. 30. Htet AT, Saturnino GB, Burnham EH, Noetscher GM, Nummenmaa A, Makarov 46. Deng ZD, Lisanby SH, Peterchev AV. Electric field depth-focality tradeoff in SN. Comparative performance of the finite element method and the bound - transcranial magnetic stimulation: simulation comparison of 50 coil designs. ary element fast multipole method for problems mimicking transcranial Brain Stimul. 2013;6:1–13. magnetic stimulation ( TMS). J Neural Eng. 2019;16:024001. 47. Roth Y, Amir A, Levkovitz Y, Zangen A. Three-dimensional distribution of the 31. Saturnino GB, Puonti O, Nielsen JD, Antonenko D, Madsen KH, Thielscher A. electric field induced in the brain by transcranial magnetic stimulation using SimNIBS 2.1: a comprehensive pipeline for individualized electric field model - figure-8 and deep H-coils. J Clin Neurophysiol. 2007;24:31–8. ling for transcranial brain stimulation. Brain and Human Body Modeling. 48. Thielscher A, Opitz A, Windhoff M. Impact of the gyral geometry on the 2019;3–25. electric field induced by transcranial magnetic stimulation. NeuroImage. 32. Murphy SC, Palmer LM, Nyffeler T, Müri RM, Larkum ME. Transcranial magnetic 2011;54:234–43. stimulation ( TMS) inhibits cortical dendrites. Elife. 2016;5:e13598. 49. Minjoli S, Saturnino GB, Blicher JU, Stagg CJ, Siebner HR, Antunes A, 33. Aberra AS, Wang B, Grill WM, Peterchev AV. Simulation of transcranial Thielscher A. The impact of large structural brain changes in chronic stroke magnetic stimulation in head model with morphologically-realistic cortical patients on the electric field caused by transcranial brain stimulation. Neuro - neurons. Brain Stimul. 2020;13:175–89. image Clin. 2017;15:106–17. 34. Cash RFH, Zalesky A, Thomson RH, Tian Y, Cocchi L, Fitzgerald PB. Subgenual 50. Bungert A, Antunes A, Espenhahn S, Thielscher A. Where does TMS stimulate Functional Connectivity predicts antidepressant treatment response to the motor cortex? Combining electrophysiological measurements and Transcranial magnetic stimulation: independent validation and evaluation of realistic field estimates to reveal the affected cortex position. Cereb Cortex. personalization. Biol Psychiatry. 2019;86:e5–e7. 2017;27:5083–94. 35. Hopman HJ, Chan SMS, Chu WCW, Lu H, Tse CY, Chau SWH, Lam LCW, 51. Kudela P, Anderson WS. Impact of gyral geometry on cortical responses to Mak ADP, Neggers SFW. Personalized prediction of transcranial magnetic surface electrical stimulation: insights from experimental and modeling stud- stimulation clinical response in patients with treatment-refractory depres- ies. J Neural Eng. 2021;18:0460e7. sion using neuroimaging biomarkers and machine learning. J Aec ff t Disord. 52. Amunts K, Zilles K. Architectonic mapping of the human brain beyond Brod- 2021;290:261–71. mann. Neuron. 2015;88:1086–107. 36. Fox MD, Liu H, Pascual-Leone A. Identification of reproducible individualized 53. Lu H. Radiomics-informed modeling for transcranial ultrasound stimulation: targets for treatment of depression with TMS based on intrinsic connectivity. Age matters. Front Neurosci. 2022;924. NeuroImage. 2013;66:151–60. 54. Sarica C, Nankoo JF, Fomenko A, Grippe TC, Yamamoto K, Samuel N, Chen R. 37. Cardenas VA, Bhat JV, Horwege AM, Ehrlich TJ, Lavacot J, Mathalon DH, Glover Human Studies of Transcranial Ultrasound neuromodulation: a systematic GH, Roach BJ, Badran BW, Forman SD, George MS, Thase ME, Yesavage JA, review of effectiveness and safety. Brain Stimul. 2022;15:737–46. Yurgelun-Todd D, Rosen AC. Anatomical and fMRI-network comparison of multiple DLPFC targeting strategies for repetitive transcranial magnetic stimulation treatment of depression. Brain Stimul. 2022;15:63–72. Publisher’s note 38. Boggio PS, Fregni F, Bermpohl F, Mansur CG, Rosa M, Rumi DO, Barbosa Springer Nature remains neutral with regard to jurisdictional claims in ER, Odebrecht Rosa M, Pascual-Leone A, Rigonatti SP, Marcolin MA, Araujo published maps and institutional affiliations. Silva MT. Eec ff t of repetitive TMS and fluoxetine on cognitive function in patients with Parkinson’s disease and concurrent depression. Mov Disord. 2005;20:1178–84.
Journal
BMC Neuroscience – Springer Journals
Published: Mar 29, 2023
Keywords: Parkinson’s disease; Transcranial magnetic stimulation; Scalp-to-cortex distance; DLPFC; Head model; Simulation