One placement, many fields: robust voxel-wise mapping with multichannel TMS

D-optimal multichannel TMS (mTMS) current sequences produce substantially more dissimilar voxel-wise cortical E-fields than random currents, achieving ≤6 mm median mapping error at the highest noise.

• Why it matters: Voxel-wise structure–function mapping needs distinct E-fields across pulses, but single-coil TMS (sTMS) relies on physical coil repositioning that is constrained by geometry and can be less robust under noisy responses.
• What they did: The authors optimized continuous mTMS currents using determinant-based D-optimal design and benchmarked them in virtual motor mapping across multiple head models, testing 5-channel planar, 6- and 12-channel spherical arrays, varying pulse counts, noise, and depth.
• The result: At the highest noise, optimized sTMS failed to converge after 100 pulses (~20–22 mm error), while all mTMS arrays reached median errors at or below 6 mm, recovered deep targets better (up to 35 mm vs 25 mm), and a magnetic-vector-potential coil-space scheme matched subject-specific optimization at 100 pulses.