Adaptive Ocular Filter Stability During EEG Noise Spikes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Adaptive filter algorithms in electroencephalographs (EEGs) become unstable due to prolonged high amplitude noise spikes in ocular sensor channels, particularly in mobile environments, leading to corruption of EEG signals and rendering them useless.
Innovation Solution
Implementing a method to control the stability of adaptive filters by calculating a stability factor, such as z-scores, and suspending the adaptive filter algorithm operation when stability factors exceed a threshold, ensuring the algorithm does not become unstable during high noise spikes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If adaptive filter algorithms are used to filter ocular artifacts from EEG signals, then the quality of EEG signals is improved, but the algorithm becomes unstable in the presence of prolonged high amplitude noise spikes
Solution Approach 1:
The patent applies preliminary action by calculating a stability factor (such as z-score) before the adaptive filter algorithm processes the signal. When the stability factor exceeds a threshold indicating potential instability, the algorithm suspends operation in advance, preventing corruption of the EEG signal while maintaining the ability to filter artifacts when conditions are stable
Solution Approach 2:
The patent introduces an intermediary stability assessment mechanism that mediates between the adaptive filter algorithm and the noisy ocular sensor signals. The stability factor calculation acts as an intermediary layer that determines whether the adaptive filter should operate, thereby protecting the system from high amplitude noise spikes while allowing normal filtering operations to proceed
2Productivity
If adaptive filter algorithms operate continuously to remove ocular artifacts, then signal cleaning is maximized, but non-ocular noise spikes cause prolonged instability
Solution Approach 1:
The patent implements periodic action by having the adaptive filter algorithm operate intermittently rather than continuously. The stability factor is calculated continuously, and the adaptive filter is activated only during periods when the stability factor indicates safe operating conditions, creating a periodic on/off pattern that maintains signal cleaning while avoiding instability during noisy periods
Solution Approach 2:
The patent applies feedback by continuously monitoring the stability factor and using this information to control the operation of the adaptive filter algorithm. When the stability factor exceeds thresholds indicating potential instability, the system provides feedback to suspend the algorithm, and when stability returns to acceptable levels, the algorithm is reactivated, creating a closed-loop control system
Data Source
AI summary
An electroencephalograph system and method for controlling the stability of an adaptive filter during high noise spikes in ocular sensor channels. The method comprises receiving a signal from at least one sensor and determining when an adaptive filter algorithm is subject to becoming unstable based on a signal from the at least one sensor. Operation of the adaptive filter algorithm is suspended while the algorithm is subject to becoming unstable.