RGBD-Based Head Models Enable Precise MRI-Free Robotic Transcranial Magnetic Stimulation

Area of Science:

• Neuroscience
• Robotics
• Medical Imaging

Background:

• Robotic transcranial magnetic stimulation (TMS) provides precise motor mapping but relies on MRI for neuronavigation, limiting its accessibility.
• Current neuronavigation systems often require structural MRI, posing logistical, financial, and safety challenges.

Purpose of the Study:

• To evaluate the feasibility of using head models generated by a depth (RGBD) camera as an MRI-free alternative for robotic TMS.
• To assess the geometric accuracy and functional reliability of RGBD-based head models compared to MRI-based models for robotic TMS.

Main Methods:

• A within-subject design was employed with healthy participants undergoing robotic motor mapping.
• Individual head models were constructed using both MRI and RGBD camera data.
• Geometric registration accuracy and test-retest reliability of resting motor threshold (RMT), map area, and hotspot coordinates were evaluated.

Main Results:

• RGBD-based head registration achieved a mean error of 2.53 ± 0.86 mm, comparable to conventional neuronavigation systems (2-4 mm).
• RGBD-guided mapping demonstrated functional reliability similar to MRI-guided mapping.
• Differences in RMT (<1% max stimulator output) and hotspot location (<1 mm) were within acceptable physiological and technical limits.

Conclusions:

• RGBD-based head models offer sufficient anatomical and functional fidelity to replace MRI in robotic TMS.
• This MRI-free workflow significantly reduces barriers to precise, automated brain mapping, enhancing clinical and research applications, especially where MRI is unavailable.