Abstract
Purpose :
The complicated anatomic of the optic nerve makes real-time monitoring and modulation challenging, and current techniques are inadequate. In this study, we developed and implanted a wireless brain-machine interface (BMI) in a goat model, enabling convenient access to the skull base for real-time monitoring and electrical stimulation of optic nerve activity. Additionally, we built computational simulation models using Comsol for safe and effective electrical stimulation of the optic nerve.
Methods :
The BMI was implanted beneath the optic chiasm via minimally invasive trans-nasal endoscopy, creating an artificial sphenoid sinus. It successfully recorded optic nerve signals and delivered electrical stimulation simultaneously. Optic chiasmatic potential (OCP) in goat was recorded by BMI during free moving, regular eat/rest, and under isoflurane anesthesia. The BMI was capable of delivering editable waveforms and pulses ranging from 0-10 V or 0-60 mA. The computational models were developed based on MRI images and detailed into different tissue types with their conductivities.
Results :
At 3 months post-implantation, skull CT scans and ophthalmic tests did not reveal any abnormalities. Compared to visual evoked potential (VEP) recordings at occipital bone, the OCP exhibited significantly stronger amplitudes, greater sensitivity to weak light stimuli and its subtle changes, and higher repeatability to light stimuli. The implanted BMI elicited conductive potentials along neural fibers at optic chiasm using relatively low and safe electrical stimulation compared to superficial electrodes like cornea. The computational simulation models demonstrated the temporal and spatial distribution of electric field and current throughout the head, as well as heat accumulation.
Conclusions :
This study successfully implanted a skull-base BMI trans-nasally with long-term safety and minimal surgical invasion. It provides a novel platform for strong-signal, sensitive, stable monitoring, as wall as effective electrical stimulation of optic nerve. Combining with the computational models, these tools will be valuable for neurophthalmology research.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.