A 4.8-μV_rms-Noise CMOS-Microelectrode Array With Density-Scalable Active Readout Pixels via Disaggregated...

Area of Science:

• Neuroscience
• Electrical Engineering
• Materials Science

Background:

• Microelectrode arrays (MEAs) are crucial for recording neural activity.
• Existing MEAs face challenges with noise and spatial resolution.
• Active-pixel sensor (APS) technology offers potential for improved MEA performance.

Purpose of the Study:

• To demonstrate a low-noise MEA using APS readout.
• To investigate the impact of disaggregated differential amplifier implementation on MEA performance.
• To assess the feasibility of high-spatial resolution neural activity mapping.

Main Methods:

• Designed and fabricated a complementary-metal-oxide-semiconductor (CMOS) active-pixel sensor-based MEA.
• Implemented a disaggregated differential amplifier circuit, separating elements into readout pixels, reference pixels, and column circuits.
• Utilized a prototype chip with 432 electrodes for neuron signal recording.

Main Results:

• Achieved a low noise level of 4.8 μV_rms.
• The readout pixel area was reduced to less than 81 μm² due to circuit disaggregation.
• Successfully observed neuron signals of approximately 100 μV.
• Demonstrated technological feasibility for 12-μm-pitch electrode density and 6,912 readout channels.

Conclusions:

• The developed APS-based MEA offers a significant reduction in noise.
• The disaggregated amplifier design enables high-density electrode arrays and a large number of channels.
• This technology holds promise for high-spatial resolution mapping of neural network activity.