Dataset Information

Machine learning and individual variability in electric field characteristics predict tDCS treatment response.

ABSTRACT: Background: Transcranial direct current stimulation (tDCS) is widely investigated as a therapeutic tool to enhance cognitive function in older adults with and without neurodegenerative disease. Prior research demonstrates that electric current delivery to the brain can vary significantly across individuals. Quantification of this variability could enable person-specific optimization of tDCS outcomes. This pilot study used machine learning and MRI-derived electric field models to predict working memory improvements as a proof of concept for precision cognitive intervention.

Methods: Fourteen healthy older adults received 20 minutes of 2 mA tDCS stimulation (F3/F4) during a two-week cognitive training intervention. Participants performed an N-back working memory task pre-/post-intervention. MRI-derived current models were passed through a linear Support Vector Machine (SVM) learning algorithm to characterize crucial tDCS current components (intensity and direction) that induced working memory improvements in tDCS responders versus non-responders.

Main results: SVM models of tDCS current components had 86% overall accuracy in classifying treatment responders vs. non-responders, with current intensity producing the best overall model differentiating changes in working memory performance. Median current intensity and direction in brain regions near the electrodes were positively related to intervention responses (r=0.811,p<0.001 and r=0.774,p=0.001).

Conclusions: This study provides the first evidence that pattern recognition analyses of MRI-derived tDCS current models can provide individual prognostic classification of tDCS treatment response with 86% accuracy. Individual differences in current intensity and direction play important roles in determining treatment response to tDCS. These findings provide important insights into mechanisms of tDCS response as well as proof of concept for future precision dosing models of tDCS intervention.

SUBMITTER: Albizu A

PROVIDER: S-EPMC7731513 | biostudies-literature | 2020 Nov - Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning and individual variability in electric field characteristics predict tDCS treatment response.

Albizu Alejandro A Fang Ruogu R Indahlastari Aprinda A O'Shea Andrew A Stolte Skylar E SE See Kyle B KB Boutzoukas Emanuel M EM Kraft Jessica N JN Nissim Nicole R NR Woods Adam J AJ

Brain stimulation 20201010 6

<h4>Background</h4>Transcranial direct current stimulation (tDCS) is widely investigated as a therapeutic tool to enhance cognitive function in older adults with and without neurodegenerative disease. Prior research demonstrates that electric current delivery to the brain can vary significantly across individuals. Quantification of this variability could enable person-specific optimization of tDCS outcomes. This pilot study used machine learning and MRI-derived electric field models to predict w ...[more]

PMID: 33049412

Similar Datasets

Project description:BackgroundUnique amongst brain stimulation tools, transcranial direct current stimulation (tDCS) currently lacks an easy or widely implemented method for individualizing dosage.ObjectiveWe developed a method of reverse-calculating electric-field (E-field) models based on Magnetic Resonance Imaging (MRI) scans that can estimate individualized tDCS dose. We also evaluated an MRI-free method of individualizing tDCS dose by measuring transcranial magnetic stimulation (TMS) motor threshold (MT) and single pulse, suprathreshold transcranial electrical stimulation (TES) MT and regressing it against E-field modeling. Key assumptions of reverse-calculation E-field modeling, including the size of region of interest (ROI) analysis and the linearity of multiple E-field models were also tested.MethodsIn 29 healthy adults, we acquired TMS MT, TES MT, and anatomical T1-weighted MPRAGE MRI scans with a fiducial marking the motor hotspot. We then computed a "reverse-calculated tDCS dose" of tDCS applied at the scalp needed to cause a 1.00 V/m E-field at the cortex. Finally, we examined whether the predicted E-field values correlated with each participant's measured TMS MT or TES MT.ResultsWe were able to determine a reverse-calculated tDCS dose for each participant using a 5 × 5 x 5 voxel grid region of interest (ROI) approach (average = 6.03 mA, SD = 1.44 mA, range = 3.75-9.74 mA). The Transcranial Electrical Stimulation MT, but not the Transcranial Magnetic Stimulation MT, significantly correlated with the ROI-based reverse-calculated tDCS dose determined by E-field modeling (R2 = 0.45, p < 0.001).ConclusionsReverse-calculation E-field modeling, alone or regressed against TES MT, shows promise as a method to individualize tDCS dose. The large range of the reverse-calculated tDCS doses between subjects underscores the likely need to individualize tDCS dose. Future research should further examine the use of TES MT to individually dose tDCS as an MRI-free method of dosing tDCS.

Dataset Information

Machine learning and individual variability in electric field characteristics predict tDCS treatment response.

Publications

Machine learning and individual variability in electric field characteristics predict tDCS treatment response.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets