A Modular 512-Channel Neural Signal Acquisition ASIC for High-Density 4096 Channel Electrophysiology.

Area of Science:

• Neuroscience
• Electrical Engineering
• Biomedical Engineering

Background:

• Brain information processing demands high spatial and temporal resolution.
• Dense electrode arrays and high-channel count electronics are crucial for advanced neural recording.
• Existing technologies face limitations in scalability and noise performance.

Purpose of the Study:

• To develop an ultra-low noise, scalable neural recording circuit.
• To create a flexible, high-channel count module for brain-computer interfaces.
• To demonstrate the capability for in vivo neural signal acquisition.

Main Methods:

• Designed a 512-channel neural readout application-specific integrated circuit (ASIC) with a 2D layout.
• Fabricated the chip using TSMC 0.18 µm 1.8 V CMOS technology.
• Integrated the ASIC into an ultra-light, flexible module with programmable analog front-ends and 14-bit ADCs.

Main Results:

• Achieved ultra-low noise performance with 512 simultaneously recording channels, scalable to 4096.
• Demonstrated a flexible, lightweight module (350 mg) suitable for small animal headstages.
• Successfully recorded neural signals, including spikes and field potentials, in vivo.

Conclusions:

• The developed neural recording circuit offers a scalable and high-performance solution for dense neural interfacing.
• The flexible module facilitates advanced neuroscience research by enabling high-resolution brain activity monitoring.
• The technology shows promise for applications requiring precise measurement of neural dynamics.