Neural Recording Front End Using Transconductance Multiplexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current neural recording technologies face challenges in efficiently amplifying and converting the small amplitude electrical signals from neural electrodes while minimizing memory effects and crosstalk, which can lead to signal-to-noise ratio drops and increased complexity in processing multi-channel signals.
Innovation Solution
The proposed neural recording apparatus employs transconductance circuits with transistors and capacitors to remove direct current components and generate current signals, which are then multiplexed and converted into digital signals using an analog-to-digital converter, reducing memory effects and crosstalk through a regulator and multiplexer configuration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If typical analog front-end circuits with instrumentation amplifier and ADC are used for each electrode, then neural signals can be amplified and converted, but memory effects and crosstalk increase leading to signal-to-noise ratio drops
Solution Approach 1:
The patent divides the neural recording system into multiple independent channels, each with its own instrumentation amplifier and ADC. This segmentation isolates signal processing paths to prevent crosstalk between channels while maintaining individual signal integrity for accurate neural recording
Solution Approach 2:
The patent introduces dedicated buffer amplifiers and impedance matching circuits as intermediary elements between electrodes and the main signal processing path. These intermediaries isolate the high-impedance electrode signals from the low-impedance amplifier inputs, preventing memory effects and signal degradation
2Quantity of substance
If multiple channels are processed simultaneously, then comprehensive neural data is captured, but processing complexity increases
Solution Approach 1:
The patent implements independent processing chains for each channel with dedicated amplifiers, filters, and ADCs. This segmentation allows parallel processing of multiple channels without mutual interference, capturing comprehensive neural data while keeping each processing path simple and manageable
Solution Approach 2:
The patent employs universal circuit blocks that can be replicated across multiple channels, such as standardized instrumentation amplifier designs and identical ADC interfaces. This modular universality reduces overall system complexity by using repeated simple units rather than complex custom designs for each channel
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for efficient amplification and conversion of neural signals, reducing memory effects and crosstalk, thereby improving signal-to-noise ratios and simplifying processing, while maintaining a compact form factor suitable for integration with microelectrode arrays.
Implementation Method 1
transconductance circuits configured to generate respective current signals by performing transconductance based on multiple voltage signals
Implementation Method 2
the capacitor of each of the transconductance circuits may be configured to remove a direct current (DC) component of a respectively received voltage signal
Data Source
AI summary
A neural recording apparatus including an electrode array including a plurality of electrodes configured to detect voltage signals of one or more neurons and a reference electrode configured to detect a reference signal, a regulator configured to regulate the reference signal, a plurality of transconductance circuits configured to generate current signals by performing transconductance based on the voltage signals and the regulated reference signal, a multiplexer (MUX) configured to multiplex on the generated current signals, and an analog-to-digital converter (ADC) configured to convert the multiplexed current signals into a digital signal