Transcranial Magnetic Stimulation is a non-invasive clinical therapy used to treat depression and migraine, and shows further promise as treatment for Parkinson’s disease, Alzheimer’s disease, and other neurological disorders. However, it is yet unclear as to how anatomical differences may affect stimulation from this treatment. We use finite element analysis to model and analyze the results of Transcranial Magnetic Stimulation in various head models. A number of heterogeneous head models have been developed using MRI data of real patients, including healthy individuals as well as patients of Parkinson’s disease. Simulations of Transcranial Magnetic Stimulation performed on 22 anatomically different models highlight the differences in induced stimulation. A standard Figure of 8 coil is used with frequency 2.5 kHz, placed 5 mm above the head. We compare cortical stimulation, volume of brain tissue stimulated, specificity, and maximum E-field induced in the brain for models ranging from ages 20 to 60. Results show that stimulation varies drastically between patients of the same age and health status depending upon brain-scalp distance, which is not necessarily a linear progression with age.

Transcranial Magnetic Stimulation (TMS) uses external time-varying magnetic fields to induce electric fields within the brain to downregulate or upregulate targeted brain tissue. Although TMS is currently only FDA approved to treat depression and migraine, it is also a promising treatment for certain symptoms of Parkinson’s disease (PD), Schizophrenia, Obsessive Compulsive disorder, and other neurological conditions.1,2 We use data obtained from the Human Connectome Project (HCP), a study sponsored by the National Institute of Health which gathered a large amount of Magnetic Resonance Images (MRI) from real individuals.3 In previous studies, we have shown Transcranial Magnetic Stimulation induced in models developed using data from the HCP.4 Here, we introduce models of Parkinson’s patients created using data obtained from the Parkinson’s Progression Marker’s Initiative (PPMI), a project which collected MRI images of patients with Parkinson’s disease.5 We created head models of PPMI patients with the goal of comparing effects of TMS on healthy individuals and those with Parkinson’s disease. It is well known that the structure of the brain in Parkinson’s patients can vary drastically, depending on disease progression, due to lesions which form on white matter surface.6,7 Furthermore, it is well known that brain volume generally decreases with age as the brain shrinks, but this fact is variable with genetics and lifestyle.8 We assert that these structural difference may influence the stimulation of the brain during TMS. To study these effects and draw relationships between brain structure and stimulation effects, we compare results of TMS simulations on models of healthy individuals (henceforth referred to as HCP models) and Parkinson’s patients (PPMI models) of varying ages. We also include the highly heterogeneous models known as Duke and Ella, developed by the IT’IS Foundation using MRI data of healthy individuals. We compute maximum E-field (Emax) induced in the brain as well as total volume and surface of the brain which receives stimulation as well as the specificity of the induced E-field. We compare these effects for patients of varying ages and discuss the implications for brain stimulation.

Simulations include commercial TMS Figure-of-8 coil run using current of 5000 Amps at a frequency of 2.5 kHz. The TMS coils are situated 5 mm above the surface of the head in each case, as shown in Figure 1. To find optimum power required to stimulate each patient’s brain tissue, clinicians find the motor cortex area of the brain and induce enough stimulation to obtain a visible motor reaction (twitching of the hand), and this is called the motor threshold (MT). 120% of MT is then used as the stimulation parameter in the targeted region of the brain. To mimic clinical TMS, we define our stimulation threshold as half of the maximum E-field induced in a particular model, which is close to MT for a healthy individual. Because Emax in each model varies, we also define an absolute stimulation threshold of 50 V/m for purposes of comparison. Then, we analyze the amount of brain volume stimulated, surface stimulated, and specificity, a measure of how focused the E-field diffusion is in the brain tissue. We compute specificity by dividing volume stimulated by surface stimulated. We show the results of both nominal stimulation, which varies for each model, as well as stimulation above the absolute threshold of 50 V/m, for purposes of clarity and comparison. Surface stimulation and specificity do not include the Duke and Ella models due to IT’IS licensing restrictions. Biot-Savart Law and Maxwell’s equations are used to calculate H-field, B-field, and E-field induced in the tissues, similar to our previous publications.9,10,11

FIG. 1.

E-field induced on surface of two HCP models (top) and two PPMI models (bottom).

FIG. 1.

E-field induced on surface of two HCP models (top) and two PPMI models (bottom).

Close modal

Figure 2 shows electric field induced in two HCP models and two PPMI models. We compare the induced electric field by analyzing the number of cells which receive stimulation above the given threshold, both throughout the brain (volume stimulated), and on the surface of the grey matter (surface stimulated). Stimulation threshold is first taken as Emax/2 and then as 50 V/m, for separate comparisons.

FIG. 2.

Two models demonstrating orientation of TMS coils.

FIG. 2.

Two models demonstrating orientation of TMS coils.

Close modal

A significant cause of stimulation variability is the distance between the brain and the scalp, which increases with age. It is unclear how Parkinson’s disease affects this process, as it is possible that the disease may cause structural differences in brain tissue.6 Figure 3 shows how the brain-scalp distance varies with age in our models. The trend is much clearer with the younger healthy individuals than for Parkinson’s patients. For this reason, we will compare all parameters using brain-scalp distance, rather than age, as the independent variable. Figure 4 shows the maximum E-field values induced in all models as brain-scalp distance varies. It is clear that in general, as the brain-scalp distance increases, a lower maximum E-field value is induced in the brain tissue, in both HCP and PPMI models. Because Emax varies for each model, and our stimulation threshold is generally dependent on Emax, it is important to also choose an unconditional threshold to decipher any trends between stimulation and brain-scalp distance. We take this absolute threshold as 50 V/m, because each model receives at least that value of stimulation. When this absolute threshold is used, we use the phrase “absolute stimulation”. Figure 5 shows volume and surface stimulation in each model as a function of brain-scalp distance. Relationships are much more obvious in the plots utilizing the absolute threshold of 50 V/m. It is shown that as the brain-scalp distance increases, volume stimulation and surface stimulation both decrease, which is intuitive given that the magnetic field decays as 1/r3, and any increase in the distance between the brain and the coil would reduce the induced E-field significantly. Furthermore, we calculate specificity (volume stimulated divided by surface stimulated) as a function of distance (Fig 6(a)), as well as a function of Emax (Fig 6(b)). Specificity as a function of distance does not show much of a trend, once again due to the variability of Emax in each model. For this reason, then, we consider specificity as a function of Emax, and discover a clear monotonically increasing relationship which seems to differ between HCP and PPMI models. Models with high maximum E-field values also receive higher specificity of stimulation.

FIG. 3.

Brain-Scalp Distance as a function of age, with blue data points representing healthy individuals and red data points indicating Parkinson’s patients.

FIG. 3.

Brain-Scalp Distance as a function of age, with blue data points representing healthy individuals and red data points indicating Parkinson’s patients.

Close modal
FIG. 4.

Maximum E-field values induced in each model as a function of brain-scalp distance.

FIG. 4.

Maximum E-field values induced in each model as a function of brain-scalp distance.

Close modal
FIG. 5.

Volume of brain tissue and grey matter surface stimulated as a function of brain-scalp distance. 5a and 5b show volume stimulated, 5c and 5d show surface stimulated.

FIG. 5.

Volume of brain tissue and grey matter surface stimulated as a function of brain-scalp distance. 5a and 5b show volume stimulated, 5c and 5d show surface stimulated.

Close modal
FIG. 6.

Specificity with varying brain-scalp distance (a) and Emax (b).

FIG. 6.

Specificity with varying brain-scalp distance (a) and Emax (b).

Close modal

While the volume, surface, and specificity vary drastically with between models, we are able to discern clear trends when we observe these same parameters using the constant stimulation threshold of 50 V/m. While general trends have been observed in the past between the patient’s age and brain-scalp distance, these trends can differ drastically between healthy individuals and Parkinson’s patients. In PPMI models, the variability in brain-scalp distance was much greater than in the HCP models. This may be due to the disease or to age, both of which can alter brain structure.6,7 Furthermore, maximum induced E-field can vary between patients of the same age. As Emax increases, the specificity is monotonically increasing for both HCP and PPMI models. From these data, it is clear that stimulation effects can vary drastically between patients of similar age and health status. The causes for this discrepancy may be numerous, but we can infer from our results that stimulation effects are more far more dependent upon brain-scalp distance than upon age and health. This implies that in order to accurately determine TMS parameters for a particular patient, it is crucial to determine the brain-scalp distance using MRI.

Data were provided [in part] by the Human Connectome Project, WU-Minn Consortium funded by the 16 NIH Institutes and Centers that support the NIH Blueprint for Neuroscience Research; and by the McDonnell Center for Systems Neuroscience at Washington University, and from the Parkinson’s Progression Marker’s Initiative (PPMI) database (www.ppmi-info.org/data). We are grateful to Gabrielle Briana Marie Jones for helpful comments.

1.
E. M.
Wassermann
and
S. H.
Lisanby
, “
Therapeutic application of repetitive transcranial magnetic stimulation: A review
,”
Clin. Neurophysiol.
112
,
1367
1377
(
2001
).
2.
A.
Pascual-Leone
, “
Transcranial magnetic stimulation: applications in basic neuroscience and neuropsychopharmacology
,”
Int. J. Neuropsychopharmacol
3
(
3
),
259
273
(
2000
).
3.
“Human Connectome Project” [Online]. Available: www.humanconnectome.org.
4.
E. G.
Lee
,
W.
Duffy
,
R. L.
Hadimani
,
M.
Waris
,
W.
Siddiqui
,
F.
Islam
,
M.
Rajamani
,
R.
Nathan
, and
D. C.
Jiles
, “
Investigational effect of brain-scalp distance on the efficacy of transcranial magnetic stimulation treatment in depression
,”
IEEE Trans. Magn.
52
,
7
(
2016
).
5.
“Parkinson’s Progression Markers Initiative,” [Online]. Available: www.ppmi-info.org.
6.
H.
Braak
,
K.
Del Tredici
,
U.
Rüb
,
R. A. I.
De Vos
,
E. N. H.
Jansen Steur
, and
E.
Braak
, “
Staging of brain pathology related to sporadic Parkinson’s disease
,”
Neurobiol. Aging
24
(
2
),
197
211
(
2003
).
7.
N. I.
Bohnen
and
R. L.
Albin
, “
White matter lesions in Parkinson disease
,”
Nat. Rev. Neurol.
7
(
4
),
229
236
(
2011
).
8.
R. C.
Gur
,
P. D.
Mozley
,
S. M.
Resnick
,
G. L.
Gottlieb
,
M.
Kohn
,
R.
Zimmerman
,
G.
Herman
,
S.
Atlas
,
R.
Grossman
, and
D.
Berretta
, “
Gender differences in age effect on brain atrophy measured by magnetic resonance imaging
,”
Proc. Natl. Acad. Sci. U. S. A.
88
(
7
),
2845
2849
(
1991
).
9.
Sim4Life, “
Simulation Theory
,” in Sim4Life 3.0 Manual, Zurich, Switzerland,
Zurich MedTech
,
2016
.
10.
L. J.
Crowther
,
R. L.
Hadimani
,
A. G.
Kanthasamy
, and
D. C.
Jiles
, “
Transcranial magnetic stimulation of mouse brain using high-resolution anatomical models
,”
Journal of Applied Physics
115
(
17
) (
2014
).
11.
L. J.
Crowther
,
R. L.
Hadimani
, and
D. C.
Jiles
, “
Effect of anatomical brain development on induced electric fields during transcranial magnetic stimulation
,”
IEEE Trans. Magn.
50
(
11
) (
2014
).