On the importance of precise electrode placement for targeted transcranial electric stimulation
Introduction
Transcranial electric stimulation (TES), including transcranial direct current (tDCS) or alternating current (tACS) stimulation, is an increasingly popular method for non-invasive modulation of neural activity in humans (Paulus, 2011). Typically, weak electric currents (e.g. 1 mA) are passed through two or more electrodes attached to the scalp creating a low amplitude electric field in the brain. Repeat administration of these currents is increasingly considered as a potential therapeutic modality for psychiatry due to the ability to produce sustained changes in neural function, possibly by inducing neuroplastic changes (Kuo et al., 2014). As TES moves towards the clinical realm, the need for consistent, reliable administration of TES across sessions and individuals becomes increasingly important.
A common practice in the application of TES is to equate the placement of electrodes across individuals using anatomical landmarks defined using a reference system, such as the International 10–20 system (Woods et al., 2016). However, intracranial electric field measurements have shown that the spatial distribution of the electric fields (including orientation and strength) during TES can have intricate patterns (Huang et al., 2017; A. Opitz et al., 2016a), which significantly increases the difficulty of creating reliable stimulation protocols. In this regard, current practices tend to rely on consistent placement of reference systems for the identification of anatomical landmarks to guide the targeting of stimulation; however, there have been only limited efforts to establish acceptable tolerance limits for variation in placement across administrations.
Realistic computational models of the brain offer a potential solution for increasing the spatial accuracy of targeting for stimulation. In addition to accounting for the impact of the expected variations in anatomy among individuals, they provide a medium for making predictions about the influences of anatomical factors that can vary across the lifespan, or can be affected by disease processes (e.g., Alzheimer's disease). Examples of such factors include gyral folding, CSF thickness, and skull composition (Opitz et al., 2015). Additionally, they can provide insights into the impact of commonly overlooked technical factors, such as skin conductance and electrode size (Saturnino et al., 2015). Researchers are increasing the use of realistic brain models to devise electrode montages, and to interpret variations in TES outcomes within and across studies investigating differences in electric field spread and strength across individuals (Laakso et al., 2015).
Here, we leverage individual-specific realistic brain models to inform our understanding of variations in the electric field generated by differences in electrode placement from administration to administration, and generate practical guidelines for decreasing this variability. We: a) carry out a validation for the specific realistic brain modeling framework used in the present work; this work confirms the findings of an initial validation effort recently carried out in ten neurosurgical patients (Huang et al., 2017) and extends it to provide an understanding of the impact of skull defects and surgical materials on findings, and b) use the validated model to establish estimates of the tolerance limits for the placement of electrodes; tolerance is determined with respect to the consistency of the spatial distribution of the electric field and that of the electric field strength generated. This allows us to derive an estimate of the minimal accuracy needed for electrode placement to reliably administer targeted transcranial electrical stimulation.
Section snippets
Model validation
Experimental data was obtained from a 29-year-old male patient and a 35-year old female patient with refractory epilepsy who underwent presurgical monitoring at North Shore University Hospital. The experimental protocol was approved by the Institutional Review Board of the Feinstein Institute for Medical Research; the patients gave informed consent in accordance with the ethical standards of the Declaration of Helsinki and monitored by the local Institutional Review Board. Generalization to
Realistic brain model generation
To identify those factors that most impact the findings generated using realistic brain models, we created four distinct FEM head models of increasing complexity in multiple steps (see Fig. 1). First, we reconstructed WM, GM, ventricles and skin surfaces from the high-resolution pre-implantation T1 using Simnibs (Thielscher et al., 2015; Windhoff et al., 2013). The skull was segmented based on intensity thresholding and manual corrections from the co-registered CT image. Most importantly, the
Realistic head model validation
The first step in our work was to validate the finite element method approach used to simulate TES electric fields. Specifically, we compared the spatial pattern of the electric field predicted by the realistic head model with the actual electric field measurements obtained from implanted electrodes. To maximize accuracy, the model accounted for the impact of skull defects and the ECOG grid with a realistic conductivity of 10−10 S/m in the model. As depicted in Fig. 2, the realistic FEM model
Discussion
The present work leveraged realistic brain models to estimate the consistency of electric field generation across TES administrations as a function of electrode montage placement. The ability of realistic head models to meaningfully predict in vivo measurements of the electric field generated by TES administration in neurosurgical patients was found to be compromised when actual electrode placement differed by more than 1 cm. Parametric manipulation of electrode placement in realistic brain
Financial disclosures
A.O. is an inventor on patents and patent applications describing methods and devices for noninvasive brain stimulation. EY, AT, CS, AM, MM have no conflict of interest.
Acknowledgements
Supported by MH110217, MH111439, pilot funding from NKI and the University of Minnesota’s MnDRIVE (Minnesota’s Discovery, Research and Innovation Economy) Initiative. Transcranial electrical stimulation devices were obtained through an equipment award provided to NKI by the Child Mind Institute (1FD2013-1). AT received support from Lundbeckfonden (R118-A11308) and NovoNordisk fonden (NNF14OC0011413).
References (24)
- et al.
Transcranial direct current stimulation in patients with skull defects and skull plates: high-resolution computational FEM study of factors altering cortical current flow
Neuroimage
(2010) - et al.
Therapeutic effects of non-invasive brain stimulation with direct currents (tDCS) in neuropsychiatric diseases
Neuroimage
(2014) - et al.
An integrated framework for targeting functional networks via transcranial magnetic stimulation
Neuroimage
(2016) - et al.
On the importance of electrode parameters for shaping electric field patterns generated by tDCS
Neuroimage
(2015) - et al.
The Human Connectome Project: a data acquisition perspective
Neuroimage
(2012) - et al.
A technical guide to tDCS, and related non-invasive brain stimulation tools
Clin. Neurophysiology
(2016) - et al.
Combining EEG and MEG for the reconstruction of epileptic activity using a calibrated realistic volume conductor model
PLoS One
(2014) - et al.
Electrical impedance tomography
SIAM Rev.
(1999) - et al.
Cocor: a comprehensive solution for the statistical comparison of correlations
PLoS One
(2015) Statistical methods for research workers
Biol. Monogr. manuals
(1925)