Real-time electric-field neuronavigation on realistic head models for conventional and multi-locus TMS.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Medical Imaging

Background:

• Transcranial magnetic stimulation (TMS) efficacy is limited by uncertainty in targeted brain areas.
• Lack of real-time feedback leads to poor reproducibility and interpretability in TMS studies.
• Individualized stimulation accuracy is crucial for reliable TMS outcomes.

Purpose of the Study:

• To develop and validate a neuronavigation system for real-time electric field (E-field) computation and visualization.
• To extend the system's capabilities for multi-locus TMS (mTMS).
• To improve the precision, interpretability, and reproducibility of TMS procedures.

Main Methods:

• Integrated an E-field solver with InVesalius navigation software, creating a GUI for real-time E-field visualization.
• Demonstrated in vivo performance using a figure-of-eight coil and a 5-coil mTMS system.
• Quantified precision, accuracy errors, RMSE, and MAG, and validated with motor evoked potentials (MEPs).

Main Results:

• The system achieved low E-field module latency (approx. 24 ms) and rapid boundary element model generation (27 s).
• In vivo TMS showed low precision (0.9-2%) and accuracy errors (2.2-4.9%) with consistent MEP responses.
• Real-time mTMS visualization demonstrated decreased MEP amplitude as the E-field shifted away from the motor hotspot.

Conclusions:

• The developed system provides real-time, individualized E-field visualization during TMS.
• It accounts for cortical folding, enhancing the accuracy of brain stimulation.
• The system simplifies, standardizes, and improves the reproducibility of TMS procedures.