Anisotropic conductivity modeling for tDCS in Parkinson’s disease using multidimensional diffusion MRI

Published October 21, 2026. A new MRI-based model showed that predicted tDCS fields were driven more by individual anatomy than by the more complex diffusion-tensor choice. This may help focus future dose-personalization work, but no stimulation was delivered and the study does not show symptom improvement.

Osorio Jurado S, Skorpil M, Svenningsson P, Moreno R, Olsson C.
Frontiers in Human Neuroscience · Publication date: 2026-10-21
Added to site: 2026-10-22
DOI: 10.3389/fnhum.2026.1972676

Published October 21, 2026. Added to Neurofeedback.com October 22, 2026.

Modalities

tDCS, Diffusion MRI, Electric-field modeling

Clinical focus

Parkinson's disease, Individualized dosimetry, Head modeling

Population

People with Parkinson's disease assessed on medication and healthy controls

Sample

29 participants: 12 with Parkinson's disease and 17 healthy controls; all completed imaging and modeling

Intervention / method

Three conductivity models, isotropic, conventional single-shell DTI and multidimensional diffusion MRI, were compared within each participant across four modeled 2 mA motor-cortex or prefrontal montages using the same mesh, electrodes and solver

Comparator

Multidimensional diffusion MRI conductivity modeling versus conventional DTI and isotropic models; exploratory Parkinson's versus control comparison

Clinician summary

The two anisotropic models differed mainly in white-matter tensor orientation by 18 to 25 degrees, while regional modeled field magnitudes differed by no more than 6.6%. No region or montage showed a significant Parkinson's-versus-control field difference. The field-stiffness association weakened from partial r=0.58 to r=0.06-0.09 after accounting for CSF morphology.

How to interpret the findings

There was no measured ground-truth conductivity or electric field, no delivered tDCS and no clinical outcome. The Parkinson's-versus-control analysis was exploratory and underpowered. The DTI and multidimensional scans used different acquisitions, resolutions and smoothing, so orientation and shape differences cannot be attributed solely to the tensor estimator.

Equipment and disclosures

Imaging used a Philips Ingenia CX 3 T scanner with a 32-channel head coil. SimNIBS 4.6 generated the head models. The study modeled, but did not deliver, four 2 mA tDCS montages.

Imaging acquisition and preprocessing were supported by Hälsa, Medicin och Teknik, MedTechLabs, Digital Futures, Vinnova through AIDA and the Swedish Research Council; one author also received IRES travel support. Funders reportedly had no role in design, analysis, interpretation or manuscript preparation. The authors declared no commercial or financial conflicts. No correction, retraction or expression-of-concern relation appeared in the retained Crossref record; this is a bounded check. AI-assisted review.

Related evidence

Compare findings from the same clinical area, including mixed results.

• Noninvasive brain stimulation for motor function and quality of life in Parkinson’s disease: a systematic review and network meta-analysis · Positive with limitations
• Explorative investigation on effects of multi-day neurofeedback with implanted electrodes in patients with Parkinson's disease. · Mixed
• High-frequency rTMS over supplementary motor or primary motor cortex on reach-to-grasp in Parkinson's disease: A randomised controlled trial. · Positive with limitations
• Subthalamic DBS Engages Right-lateralized Frontal Control to Improve Gait Adaptation in Parkinson's Disease. · Positive with limitations