Real-time Stimulation Artifact Suppression for Neural Recording
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for recording neural signals during or after electrical stimulation are hindered by large stimulation artifacts, leading to amplifier saturation and signal loss, and struggle with motion artifacts caused by tissue deformation and electrode-tissue interface changes.
Innovation Solution
A hardware-based system that uses a square current pulse to estimate skin impedance and motion artifacts, allowing for real-time cancellation of these artifacts during simultaneous stimulation and recording, without relying on complex circuitry or sub-threshold stimulation, and enables accurate artifact template acquisition without evoking neural responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If software-based methods are used to recover the neural signal, then the neural signal can be recovered after artifact corruption, but the processing is time consuming and cannot prevent amplifier saturation
Solution Approach 1:
The system performs preliminary action by acquiring artifact templates during calibration phases before actual neural recording. The artifact templates are stored and ready for subtraction during simultaneous stimulation and recording, enabling real-time artifact removal without time-consuming post-processing.
Solution Approach 2:
The system skips the time-consuming iterative software-based artifact removal process by using pre-acquired artifact templates. The template subtraction is performed rapidly in real-time during stimulation, rushing through the artifact removal step without the delays associated with post-processing software methods.
2Measurement precision
If the neural signal amplifier provides moderate gain to amplify the neural signal, then the neural signal becomes detectable, but the amplifier becomes severely saturated due to the presence of large artifacts
Solution Approach 1:
The system applies preliminary anti-action by subtracting the artifact template from the recorded signal before amplification. This pre-subtraction removes the harmful artifact component, allowing the amplifier to gain the small neural signal without becoming saturated by the large artifact that would otherwise be present.
Solution Approach 2:
The system converts the harmful large artifact into a useful artifact template through calibration-phase acquisition. The same artifact that would cause saturation is transformed into a beneficial reference signal that, when subtracted, enables high-gain amplification of the neural signal without saturation.
3Object-affected harmful factors
If hardware approaches such as blinking and electrode discharging are used to remove artifacts, then the artifact is reduced, but signal loss occurs due to amplifier saturation or input grounding during and shortly after stimulation
Solution Approach 1:
The system performs preliminary action by acquiring the artifact template during a calibration phase before simultaneous stimulation and recording begins. This pre-acquisition allows the artifact to be characterized and stored without affecting the neural signal recording, avoiding the signal loss that occurs when artifact removal methods are applied during active recording.
Solution Approach 2:
The artifact template acts as an intermediary that mediates between the harmful artifact and the neural signal. By introducing this template signal for subtraction, the system removes the artifact without needing to apply disruptive hardware methods like blinking or electrode discharging that would cause signal loss.
4Object-affected harmful factors
If spectral cancellation of the artifact is implemented in the frequency domain, then the artifact is removed, but distortion occurs because the spectrum of the signal of interest might overlap with that of the artifact
Solution Approach 1:
The system extracts the artifact component from the recorded signal by subtracting the acquired artifact template in the time domain. This direct time-domain subtraction removes the artifact without needing to transform to the frequency domain, avoiding the distortion that would result from spectral overlap between artifact and neural signal.
Solution Approach 2:
The system substitutes the frequency-domain spectral cancellation method with a time-domain template subtraction method. This replacement of the frequency-domain approach with a time-domain approach eliminates the distortion problem caused by spectral overlap while maintaining effective artifact removal.
5Reliability
If sub-threshold stimulus is used to acquire artifact template, then neural responses are avoided, but the artifact template may not accurately represent artifacts from larger stimulus intensities
Solution Approach 1:
The system changes the stimulus intensity parameter during calibration to match the actual stimulation conditions. By acquiring the artifact template using the same stimulus intensity as the subsequent simultaneous stimulation and recording, the artifact template accurately represents the artifacts that will be present, resolving the mismatch problem of using sub-threshold intensities.
6Device complexity
If the same electrode is used for both stimulation and recording, then device complexity is reduced, but large artifacts are generated that saturate the recording device
Solution Approach 1:
The system converts the harmful large artifact generated by using the same electrode for stimulation and recording into a beneficial artifact template. By acquiring this template during calibration and subtracting it during simultaneous operation, the large artifact that would otherwise saturate the recorder becomes a useful reference that enables artifact removal and successful recording.
Data Source
AI summary
A system and method of suppressing stimulation artifacts when performing electrophysiological electrical stimulation and recording. An artifact waveform is captured associated with a stimulus output, and then the artifact waveform calibrated during another stimulus output for accurately representing the actual artifact waveform received within each measured response to a stimulus. During actual stimulus generation and response recording, the calibrated artifact waveform is subtracted in at least one of the amplifier stages so that the artifacts are removed from the amplified response to the stimulus thus providing an accurate output without saturating the amplifiers.


