Implantable Device Neural Potential Detection via Signal Variance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical devices face challenges in detecting neural potentials evoked by electrical stimulation due to their long evolution time and potential masking by stimulation artifacts, especially at high stimulation rates and limited bandwidth, making it difficult to capture and distinguish these responses.
Innovation Solution
An implantable medical device utilizes variations in electrical signals, such as standard deviation or variance, to differentiate second-order potentials from artifacts, allowing for closed-loop stimulation adjustments and improved detection of neural responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If electrical stimulation is delivered at high rates, then therapeutic efficacy is improved, but neural potentials are masked by stimulation artifacts making detection difficult
Solution Approach 1:
The patent extracts the neural potential signal from the composite signal by separating it from the stimulation artifact. This is achieved through signal processing techniques that isolate the evoked potential component from the dominant artifact, enabling detection even at high stimulation rates where the artifact would otherwise mask the neural response.
Solution Approach 2:
The patent introduces an intermediary signal processing stage between stimulation delivery and detection. This intermediary processing includes filtering, averaging, or other computational methods that enable the detection system to distinguish neural potentials from artifacts, effectively acting as a mediator that resolves the masking problem.
2Device complexity
If bandwidth is limited, then device complexity is reduced, but ability to capture and distinguish neural responses deteriorates
Solution Approach 1:
The patent applies preliminary signal processing actions before final detection and analysis. By pre-processing the signals through filtering, averaging multiple stimuli responses, or other computational preparations, the system enhances the detectability of neural potentials within limited bandwidth constraints, avoiding the need for higher bandwidth hardware.
Solution Approach 2:
The patent creates multiple copies of the stimulation signal and averages them together. By delivering repeated stimuli and averaging the responses, the system enhances the signal-to-noise ratio and makes neural potentials detectable even with limited bandwidth, as the consistent evoked potential pattern emerges from the averaged copies.
3Loss of time
If neural potentials are detected directly, then response time is reduced, but detection precision deteriorates due to masking by artifacts
Solution Approach 1:
The patent implements continuous monitoring and processing of electrical signals during and after stimulation delivery. Rather than attempting single-point detection, the system continuously acquires signals and processes them in real-time, maintaining detection capability throughout the evolution of neural potentials while progressively improving precision through ongoing signal analysis.
Solution Approach 2:
The patent employs feedback mechanisms where the detected signals are processed and used to adjust subsequent detection parameters or stimulation delivery. This feedback loop allows the system to refine its detection approach based on observed signals, improving precision over time while maintaining responsive detection of evolving neural potentials.
Data Source
AI summary
An example system includes a memory; and processing circuitry configured to: cause an implantable stimulation device to deliver a plurality of doses of electrical stimulation to a patient; receive, for each respective dose of the plurality of doses, a respective electrical signal of a plurality of electrical signals; and determine, based on a variation of the plurality of electrical signals, whether the plurality of doses of electrical stimulation evoked neural potentials in the patient.