1946

Evaluation of dual-echo EPI for in-vivo current mapping in individual subjects during transcranial Direct Current Stimulation
Mayank S Jog1, Lirong Yan2, Kay Jann2, Lucas Parra3, Marom Bikson3, and Danny JJ Wang2

1Biomedical Engineering, University of California Los Angeles, Los Angeles, CA, United States, 2University of Southern California, CA, United States, 3Biomedical Engineering, City College of New York, NY, United States

Synopsis

Transcranial Direct Current Stimulation (tDCS) is a neuromodulation technique that uses milliampere currents to modulate cortical excitability. Although tDCS has been shown to treat clinical symptoms and improve cognition, the distribution of tDCS currents in the brain remains unknown. Here we show a MRI technique that measures a component of the magnetic field induced by tDCS currents in-vivo. Experimental data acquired using this technique is compared to model-based simulations. Our results demonstrate that mapping the tDCS current-induced magnetic fields in individual subjects is feasible; opening an avenue to map electric currents directly and track target engagement in individual subjects.

INTRODUCTION:

Transcranial Direct Current stimulation (tDCS) is a neuromodulation technique that uses milliampere currents delivered through scalp electrodes. Hypothesized to modulate cortical excitability, tDCS has been shown to treat clinical symptoms as well as improve cognition1-3. To understand the mechanism of tDCS, there is an urgent need to map the distribution of tDCS electric currents in the brain. To date, mathematical models4,5 have been used for this purpose; however these lack in-vivo validation. Recently, a technique based on Ampere’s law to visualize tDCS currents in-vivo was reported6. This technique uses MRI field mapping to detect tDCS induced magnetic field changes along the static MRI field (B0). The authors measured tDCS induced magnetic fields in a group of 12 subjects. In this abstract, we present a dual-echo EPI technique sensitive enough to perform the same measurement in individual subjects. We evaluate our technique by comparing the experimentally measured field changes to simulations (the latter based on mathematical models mentioned above and utilizing information from a separate structural MRI scan).

METHODS:

tDCS Setup: A bilateral tDCS montage was used, with sponge electrodes (7x5cm) placed at C3/C4 positions on the scalp.
Data Acquisition: MRI data was collected from 4 healthy male subjects on a Siemens 3T Trio TIM MRI using a 12-channel head coil. Field mapping data was acquired using a dual-echo EPI pCASL sequence (Fig 1). Only the field mapping data was analyzed and is reported in this abstract. During each EPI scan, 4 different tDCS stimulation blocks (0/0.5/1/1.5mA) were applied in randomized order (Fig 1). A total of 3 such EPI scans were performed per subject. The order of the currents was counterbalanced across scans and subjects. During the scans, subjects were asked to fixate on a white cross on a black background. Additionally, a T1-weighted MPRAGE structural scan was also acquired (Fig 1).
Data Processing: Phase data from the three EPI acquisitions was concatenated and unwrapped using the Region-Growth algorithm implemented in PhaseTools7. MRI data was then realigned to the first volume using SPM8.
Statistical Analyses: Phase was modeled as:
Φ = Φ0 + ΦtDCS-Current + ΦMotion + ΦDrift + ΦOff-Res + ΦNoise
Φ0 was modeled out by using the phase difference between the two TE’s. ΦDrift was modeled out by detrending each scan with a quadratic polynomial, following which magnetic field was calculated using the scaling factor 1/γ*ΔTE. Subsequently, data was modeled using a voxel-wise general linear model (GLM) with the applied current as a predictor. In the GLM, ΦOff-Res is modeled out as an intercept. Nuisance regressors comprising motion parameters, scan-number and global mean phase (modeling uncorrected global drift) were also included. The slope corresponding to the predictor was interpreted as the induced magnetic field per mA of tDCS current, with a confidence given by the goodness of fit.
Simulations: SIMNIBS8 (an open-source software) was used to define tDCS electrodes on the structural data and calculate current densities (assuming standard isotropic tissue-conductivity values). Subsequently, an in-house implementation of the biot-savart law was used to convert the current densities into magnetic fields to compare with experimental data.

RESULTS:

Fig 2 shows the measured magnetic field data (1st row) with simulation results (2nd row). Experimental measurements and simulated results were found to be significantly correlated (Spearman correlations = 0.53, 0.45, 0.13, 0.61; p < 0.001, for Subjects 1-4 respectively). However, the magnetic field in the brain was measured to be larger than its simulated counterpart in all subjects. Additionally, some inter-individual variability was observed experimentally, which was not apparent through simulations (Subject 4 vs. the rest). To help interpret magnetic fields, a separate simulation was performed (Fig 3) which showed that magnetic fields inside the brain reduce as current shunts away from the brain.

DISCUSSION and CONCLUSION:

Experimental and simulated data were significantly correlated for all subjects. Taken together with the fact that simulations predict current in the brain (Fig 3), our measurements indicate that tDCS currents do indeed pass through the brain. In fact, the observation of larger-than-simulated magnetic fields in the brain suggests that more current may be flowing through the brain parenchyma than is predicted by models (since fields in the brain reduce as more current shunts over the scalp, Fig 3). Our results show that experimental measurement of tDCS induced magnetic fields is feasible in individual subjects and can be used to guide model development. Moreover, being able to measure tDCS induced fields in single subjects opens up an avenue to measure target engagement directly: a capacity which does not exist to date.

Acknowledgements

No acknowledgement found.

References

1. Gomez Palacio Schjetnan A, Faraji J, Metz G, et al. Transcranial Direct Current Stimulation in Stroke Rehabilitation: A Review of Recent Advancements. Stroke Research and Treatment 2013, 1–14 (2013).

2. Woods AJ, Antal A, Bikson M, et al. A technical guide to tDCS, and related non-invasive brain stimulation tools. Clinical Neurophysiology 127, 1031–1048 (2016).

3. Coffman B, Clark V, Parasuraman R. Battery powered thought: Enhancement of attention, learning, and memory in healthy adults using transcranial direct current stimulation. NeuroImage 85, 895–908 (2014).

4. Bikson M, Rahman A, Datta A. Computational Models of Transcranial Direct Current Stimulation. Clinical EEG and Neuroscience 43, 176–183 (2012).

5. Kessler SK, Minhas P, Woods AJ, et al. Dosage Considerations for Transcranial Direct Current Stimulation in Children: A Computational Modeling Study. PLoS ONE; 10.1371/journal.pone.0076112 (2013).

6. Jog MV, Smith RX, Jann K, et al. In-vivo Imaging of Magnetic Fields Induced by Transcranial Direct Current Stimulation (tDCS) in Human Brain using MRI. Sci. Rep. 6, 34385; doi: 10.1038/srep34385 (2016).

7. Barnhill E, Kennedy P, Hammer S, et al. Statistical mapping of the effect of knee extension on thigh muscle viscoelastic properties using magnetic resonance elastography. Physiol Meas. 2013 Dec;34(12):1675-98. doi: 10.1088/0967-3334/34/12/1675. Epub 2013 Nov 20.

8. Thielscher A, Antunes A and Saturnino GB. Field modeling for transcranial magnetic stimulation: a useful tool to understand the physiological effects of TMS. IEEE EMBS 2015, Milano, Italy

Figures

Fig 1: MRI Scan protocol: Data was acquired from 3 back to back dual-echo EPI sequences (FOV = 220 mm, 64 × 64 matrix; 24 slices, 5 mm thick, 1 mm gap; TE1/TE2 = 11/26 ms, TR = 4 s; rate-2 GRAPPA, 7/8 partial k-space; scan time = 12.5 minutes). During each scan, tDCS stimulation was applied for 2.5 minutes with 30 seconds ramp duration. The order of the currents was randomized within scan, and counterbalanced between scans and subjects. Structural data (1mm isotropic) was also acquired.

Fig 2: Comparison of tDCS current induced magnetic fields along B0 measured experimentally (Row 1), and simulated (Row 2), for 4 subjects. The simulations are a good qualitative match (Spearman correlation r = 0.53, 0.45, 0.13, 0.61; p < 0.001, for Subjects 1-4 respectively). However, the experimentally measured fields were observed to be consistently larger than their simulated counterpart. Additionally, inter-individual variability was also larger in experimental measurements than in simulations (e.g. Subject 4).

Fig 3: Simulations showing Current density distribution in Subject 1 and resultant magnetic fields induced in two cases: (a) Currents are constrained to travel through scalp and (b) using standard tissue conductivity model. Results show that the magnetic fields inside the head decrease as more current is shunted through the scalp. This suggests that if larger-than-simulated magnetic fields are observed experimentally, more tDCS current (than predicted by simulations) could be passing through the brain parenchyma. Additionally, current density map of (b) does predict tDCS currents inside the brain.

Proc. Intl. Soc. Mag. Reson. Med. 25 (2017)
1946