A CMOS Neural Interface for a Multichannel Vestibular Prosthesis

Area of Science:

• Biomedical Engineering
• Neuroscience
• Integrated Circuit Design

Background:

• Vestibular prostheses aim to restore balance and vision stabilization by modulating neural activity.
• Previous multichannel vestibular prosthesis (MVP) systems were large and power-intensive due to discrete components.

Purpose of the Study:

• To develop a compact and efficient neural-interface chip for an MVP.
• To evaluate the performance of the new ASIC-based MVP system (MVP2A) against the previous generation (MVP2).

Main Methods:

• Designed and fabricated a high-voltage CMOS application-specific integrated circuit neural interface (ASIC-NI) chip.
• Integrated the ASIC-NI with a microcontroller for biphasic stimulation pulse delivery.
• Conducted physiological tests in rhesus monkeys to measure eye movement responses.

Main Results:

• The ASIC-NI chip reduced MVP system size by 48% and power consumption by 17%.
• The ASIC-based MVP system (MVP2A) successfully elicited reflexive eye movements.
• MVP2A performance in evoking slow phase eye velocities was statistically similar to MVP2.

Conclusions:

• The developed high-voltage CMOS neural-interface chip enables a smaller and more power-efficient MVP.
• The ASIC-NI chip maintains the therapeutic efficacy of the vestibular prosthesis.
• This advancement facilitates the development of more integrated and wearable neuroprosthetic devices.