Stable, chronic in-vivo recordings from a fully wireless subdural-contained 65,536-electrode brain-computer interface...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Materials Science

Background:

• Brain-computer interfaces (BCIs) are crucial for human applications.
• Existing BCI technologies face limitations in volumetric efficiency and invasiveness.
• Advancements in miniaturization and integration are needed for next-generation BCIs.

Purpose of the Study:

• To develop a minimally invasive, high-bandwidth BCI with improved volumetric efficiency.
• To create a flexible micro-electrocorticography (μECoG) device with a high density of recording channels.
• To demonstrate chronic, reliable neural recordings and signal decoding in animal models.

Main Methods:

• Fabrication of a 50-μm-thick, flexible μECoG BCI on a CMOS substrate.
• Integration of a 256x256 electrode array (65,536 channels) with signal processing, telemetry, and wireless powering.
• Implantation below the dura in pigs and non-human primates for chronic recordings.
• Bidirectional wireless communication with an external relay station.

Main Results:

• Achieved orders-of-magnitude improvement in volumetric efficiency compared to other BCI technologies.
• Demonstrated chronic, reliable recordings for up to 2 weeks in pigs and 2 months in non-human primates.
• Successfully decoded brain signals from somatosensory, motor, and visual cortices at high spatiotemporal resolution.

Conclusions:

• The developed μECoG BCI represents a significant advancement in neural interface technology.
• The device enables minimally invasive, high-bandwidth brain signal acquisition and decoding.
• This technology holds potential for revolutionizing human applications requiring advanced brain-computer interaction.