EEG Analysis via Fixed-Order ARMA Model for Brain State Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing brain function through EEG signals lack understanding of underlying physiological mechanisms, limiting their ability to accurately measure and monitor brain function, especially in clinical settings involving alterations in consciousness.
Innovation Solution
A method using a fixed order auto-regressive moving average (ARMA) model to analyze EEG signals, where specific orders are derived from the electrocortical transfer function, allowing for the determination of complex poles that represent the brain's state, enabling the measurement of brain function and monitoring of changes induced by interventions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If statistical signal analysis methods are used to analyze EEG signals, then the analysis can be performed without knowledge of physiological mechanisms, but the measurement precision and reliability of brain function assessment is limited
Solution Approach 1:
The patent introduces an intermediary mathematical model (ARMA model with specific 8th order AR and 5th order MA parameters) that acts as a bridge between the raw EEG signals and the underlying physiological mechanisms. This model incorporates known physiological constraints and relationships, allowing the analysis to remain computationally tractable while significantly improving measurement precision by embedding domain knowledge into the analysis framework.
2Measurement precision
If complex mathematical models incorporating physiological mechanisms are used to analyze EEG signals, then the measurement precision and reliability of brain function assessment is improved, but the device complexity and computational requirements increase
Solution Approach 1:
The patent specifies fixed parameter values for the ARMA model (8th order autoregressive and 5th order moving average) that are derived from physiological considerations. By fixing these parameters rather than allowing them to vary freely, the model maintains high measurement precision while reducing computational complexity and avoiding the need for complex parameter estimation procedures during real-time analysis.
Data Source
AI summary
A method for assessing brain state by analysing mammalian brain electroencephalogram (“EEG”) recordings using an eighth order autoregressive and fifth order moving average discrete time equation.