A Whole-Head Finite Element Model for Electrical Neuromodulation via Visual Brain-Machine Interfaces

Area of Science:

• Computational neuroscience
• Biomedical engineering
• Ophthalmology

Background:

• Accurate models of the visual pathway are crucial for developing brain-machine interfaces (BMIs) for vision restoration.
• Existing models often neglect surrounding anatomical structures, limiting simulation precision.

Purpose of the Study:

• To develop a comprehensive computational model of the human head incorporating the entire visual pathway and surrounding tissues.
• To validate the model's accuracy and assess its utility in optimizing neuromodulation and BMI technologies.

Main Methods:

• Created a detailed computational model of the human head, including the eye, optic nerve, brain, orbit, and paranasal sinuses.
• Validated the model using human and large-animal electrophysiological data.
• Performed component-elimination analysis to compare the comprehensive model with simplified versions.

Main Results:

• The comprehensive model demonstrated high correlation with measured electric potentials.
• The optimized model outperformed simplified versions in accuracy.
• Identified limitations of noninvasive and safety concerns of invasive neuromodulation approaches for optic neuropathy.
• Determined transnasal stimulation at the optic chiasm as a superior stimulation site.
• Showcased potential advantages of in silico designed optic nerve prostheses over existing retinal and cortical prosthetics.

Conclusions:

• The validated, comprehensive computational model provides a versatile resource for advancing neuromodulation strategies and visual BMI technologies.
• The model facilitates the in silico design and optimization of devices for vision restoration.
• This work supports the development of safer and more effective treatments for visual impairments.