Optimized Temporal Interference for Intracranial Electrical Stimulation: Enhancing Focality and Mapping in Blind Zones
DOI: Pending DOI
Synopsis
This study evaluates the advantages of optimized Temporal Interference (TI) for intracranial electrical stimulation (IES) mapping in drug-resistant epilepsy. Using a realistic patient model (89 sEEG contacts), we simulated TI-IES across two deep-brain targets located between electrode leads. Our results demonstrate that optimized TI significantly enhances focality compared to standard optimized IES. Due to the absence of skull shielding and higher current thresholds in IES, TI-IES offers a superior intensity-focality trade-off. This approach expands mapping capabilities without additional surgical risk, providing a powerful tool for delineating complex epileptogenic networks in regions previously inaccessible to standard bipolar stimulation.
Background
Drug-resistant epilepsy often requires stereoelectroencephalography (sEEG) to localize the epileptogenic zone through intracranial recordings and stimulation mapping [1, 2, 3]. Standard depth electrical stimulation mapping (DESM) typically involves bipolar stimulation between adjacent contacts of the same electrode to localize epileptogenic nodes and eloquent regions [1, 4]. However, this approach is constrained to the immediate vicinity of the leads, leaving "blind zones" between shafts [1, 2].
Transcranial electrical stimulation (TES) optimization has evolved to maximize intensity or focality at specific targets using multi-electrode arrays [5, 6, 7, 8]. TI was recently proposed to non-invasively target deep regions by interfering high-frequency fields [9, 10]. While optimized TI has shown promise in TES [11, 12] its application in IES remains largely unexplored.
We proposed IES optimization using contacts from different electrodes (X-DESM) to stimulate regions in between leads, but the strongest field intensities are still next to the stimulating electrodes [2]. We also showed one simple empirical IES-TI stimulation demonstrating that it can overcome this limitation, but we did not use TI optimization.
This study introduces, to our knowledge, the first systematic optimization of TI-IES using a realistic patient model. We hypothesized that optimized TI offers a superior (not just better) intensity-focality trade-off for intracranial mapping than X-DESM because, unlike TES, IES fields are more intense and do not face the "electrical shield" of the skull and CSF.
Methods
We solved the forward problem using a 3D finite element method (FEM) on a ~4M element intracranial head model derived from coregistered MRI and CT images of an real epilepsy patient of the Hospital El Cruce, Buenos Aires, Argentina. Tissues were segmented and assigned isotropic conductivities [2, 8], and the 89 contacts implanted on this patient were registered using iElectrodes [13]. We computed standard IES optimization and TI-IES optimization, defining both ends of the non Region of Interest (non-ROI) constraint limits using the one-shot weighted least-squares and reciprocity solutions [8].
Optimization of standard IES followed the constrained directional maximization framework [7, 9]. For TI-IES, we implemented the non-convex "minimization-maximization" optimization to maximize the modulation depth and limit the non-ROI energy [11]. We subsampled the non-ROI constraints from ~1.5M to ~13.5k points evenly distributed at the non-ROI gray matter to speed up calculations.
We targeted in the Anterior-Posterior (AP) direction the right Insula and Hippocampus, both located at "blind zones" between electrode shafts. Although we also arbitrarily selected the AP direction for all non-ROI test points, any other orientation can be used [11], being aware that TI can have maximum MD in many directions [14].
Performance was evaluated via Intensity-Focality Pareto curves, where focality is defined as the ratio of ROI intensity (MD in the case of TI) to non-ROI energy (MD energy for TI) [8, 11]. Simulations considered current limits up to 1mA in total, a typical value used in clinics [1].
Results
Results are summarized in Figure 1, which combines intensity-focality performance curves and 3D modulation depth distributions for the Posterior Hippocampus and Insula (top and bottom rows, respectively) . The first column depicts the ROI (red) relative to the electrode positions (blue dots). Intensity-Focality trade-off curves are shown in the second column, where orange curves represent optimized TI-IES, while blue curves represent standard optimized IES. Third and fourth columns show 3D visualizations of the electric field and modulation depth intensities of a sagittal slice at the ROI’s position for representative standard IES and TI-IES solutions (highlighted with a black cross in the second column plots).
The optimization curves consistently showed that TI-IES (red) dominates standard optimized IES (blue) across all targets. For instance, for a ROI intensity of 1V/m, TI-IES achieved around two times greater focality for the Hippocampus compared to standard optimized stimulation. This advantage was more pronounced for the Insula, which is located further away from the electrodes. It can be easily observed in the 3D examples that the same intensity can be obtained with much larger focality when using optimized TI-IES instead of optimized standard IES. As expected, the most intense solution corresponds to the reciprocity maximum intensity pair solution [8, 11]. These simulations confirm that TI-IES provides a significant leap in mapping versatility without increasing surgical density.
Discussion
This study demonstrates that optimized TI-IES provides a significant technical advantage for clinical epilepsy mapping over optimized standard IES. Our findings have direct clinical relevance: TI-IES can map "blind zones" between implanted electrodes without additional leads [2]. This constitutes a major advantage in case that the electrodes were not placed at the desired ROI, a common issue due to implantation errors, anatomical constraints or just large uncertainty of the epileptogenic network location before electrode implantation.
Unlike transcranial TI, which is limited by skull shunting and scalp safety thresholds, intracranial TI leverages direct access to the brain parenchyma. Thus, the ability to use higher intensity fields allows for robust modulation depth in IES compared to TES.
Optimized TI-IES increases the effective resolution of the sEEG grid and can resolve ambiguities in the epileptogenic network when the ictal onset zone lies between leads. By integrating individualized FEM modeling and optimized TI, clinicians can prospectively plan complex stimulation patterns to test more hypotheses. This approach maximizes diagnostic yield while minimizing patient risk.
Disclosures
Authors declare no conflicts of interests related to present work. This work was supported by the UNLP EI003 grant.
References
Trébuchon A and Chauvel P 2016 Electrical stimulation for seizure induction and functional mapping in stereoelectroencephalography. J. Clin. Neurophysiol. 33 511–21
Collavini S, Fernández-Corazza M, Oddo S, Princich J P, Kochen S and Muravchik C H 2021 Improvements on spatial coverage and focality of deep brain stimulation in pre-surgical epilepsy mapping. J. Neural Eng. 18 046004
Lüders H O, Najm I, Nair D, Widdess-Walsh P and Bingman W 2006 The epileptogenic zone: general principles. Epileptic Disord. 8 1–9
Britton J W 2018 Electrical stimulation mapping with stereo-EEG electrodes. J. Clin. Neurophysiol. 35 110–4
Dmochowski J P, Datta A, Bikson M, Su Y and Parra L C 2011 Optimized multi-electrode stimulation increases focality and intensity at target. J. Neural Eng. 8 046011
Guler S, Dannhauer M, Erem B, Macleod R, Tucker D, Turovets S, Luu P, Erdogmus D and Brooks D H 2016 Optimization of focality and direction in dense electrode array transcranial direct current stimulation (tDCS). J. Neural Eng. 13 036020
Saturnino G B, Siebner H R, Thielscher A and Madsen K H 2019 Accessibility of cortical regions to focal TES: Dependence on spatial position, safety, and practical constraints. Neuroimage 203 116183
Fernández-Corazza M, Turovets S and Muravchik C H 2020 Unification of optimal targeting methods in transcranial electrical stimulation. NeuroImage 209 116403
Grossman N, Bono D, Dedic N, Kodandaramaiah S B, Rudenko A, Suk H J, Cassara A M, Neufeld E, Kuster N, Tsai L H, Pascual-Leone A and Boyden E S 2017 Noninvasive deep brain stimulation via temporally interfering electric fields. Cell 169 1029–41
Reato D, Rahman A, Bikson M and Parra L C 2013 Effects of weak transcranial alternating current stimulation on brain activity-a review of known mechanisms from animal studies. Front. Hum. Neurosci. 7 687
Huang Y, Datta A and Parra L C 2020 Optimization of interferential stimulation of the human brain with electrode arrays. J. Neural Eng. 17 036023
Brahma T, Guillen A, Moreno J, Datta A and Huang Y 2025 On the need of individually optimizing temporal interference stimulation of human brains due to inter-individual variability. Brain Stimul. 18 1373–88.
Blenkmann A O, Phillips H N, Princich J P, Rowe J B, Bekinschtein T A, Muravchik C H and Kochen S 2017 iElectrodes: a comprehensive open-source toolbox for depth and subdural grid electrode localization. Front. Neuroinform. 11 14
Fernández-Corazza M, Turovets S and Muravchik C H 2025 Closed-form expressions for the directions of maximum modulation depth in temporal interference electrical brain stimulation. J. Neural Eng. 22 056013