BCG Timing Extraction from EEG via LRM Signal Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for suppressing ballistocardiogram (BCG) artifacts in EEG recordings during MRI are inadequate due to reliance on ECG timing, which is often inaccurate and leads to poor artifact rejection, especially when ECG recordings are of poor quality or contaminated by MRI scanner noise.
Innovation Solution
A system that directly measures BCG timing from EEG recordings using a Left Mean-Right Mean (LRM) signal derived from electrodes closest to facial arteries, employing a peak detection algorithm to improve accuracy and reliability in identifying BCG events, independent of ECG timing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ECG timing is used to suppress BCG artifacts in EEG recordings, then artifact suppression can be attempted, but measurement precision of BCG timing deteriorates due to discrepancies between ECG and BCG timing
Solution Approach 1:
The patent introduces an intermediary processing step that extracts BCG timing information directly from the EEG signal itself, rather than relying on ECG timing. This intermediary extraction method uses the EEG signal's own characteristics to identify BCG events, serving as a mediator between the contaminated EEG signal and the artifact suppression process, thereby achieving more accurate timing without depending on ECG-BCG synchronization
Solution Approach 2:
The patent replaces the mechanical/electrical coupling system between ECG and BCG timing with a signal processing approach. Instead of relying on the physiological coupling mechanism that creates timing discrepancies, the invention uses digital signal processing to directly extract BCG timing from EEG, substituting the problematic timing relationship with a computational method that achieves precise timing measurement
2Loss of information
If ECG recordings are used for timing, then cardiac timing information can be obtained, but measurement precision deteriorates when ECG recordings are of poor quality or contaminated by MRI scanner noise
Solution Approach 1:
The patent enables the EEG signal to serve itself by extracting BCG timing information directly from its own characteristics rather than relying on external ECG recordings. The EEG signal contains the necessary information about BCG events, and the invention uses self-contained signal processing methods to extract this timing information, making the system self-sufficient and independent of external ECG quality
Solution Approach 2:
The patent extracts the specific BCG timing information directly from the EEG signal by isolating and analyzing the characteristic BCG waveform features within the EEG data. This extraction process separates the timing information from the contaminated EEG signal, obtaining precise BCG timing without requiring external ECG recordings, thereby solving the problem of ECG quality degradation in MRI environments
Data Source
AI summary
A system and method to extract the timing of ECG events without reliance on ECG. The system and method are based on effects of scalp pulsation (due to blood flow) on electrode location. The electrodes in the system with the closest proximity to the facial arteries are re-referenced to create a [Mean Left-Mean Right] signal (LRM). A constrained peak detection algorithm is then used to find the BCG events. Finally, an automatic error checking and correction algorithm based on inter-beat timing is applied.