Brain Network Activity Estimation Using EEG-fMRI Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating brain network activities, such as EEG and fMRI, face challenges in requiring detailed brain or skull information and involve extensive computational processing, limiting the ability to accurately and quickly estimate activities of multiple brain networks.
Innovation Solution
A brain network activity estimation system that combines EEG and fMRI data to construct a feature estimation model, calculating feature values for modules and brain networks, and determining parameters to correlate EEG signals with fMRI images, allowing quick estimation of various brain networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If inverse problem analysis of EEG measurement results is performed to obtain information for estimating brain network activities, then brain network activity estimation is achieved, but detailed information on brain shape, skull shape, or sensor position is required and large amount of operational processing is needed, leading to poor utility and large temporal granularity
Solution Approach 1:
The patent introduces fMRI measurement data as an intermediary to bridge EEG data and brain network activities. Instead of directly analyzing EEG data through complex inverse problem analysis, the system uses fMRI data (which directly shows brain network activities) as a mediator to train a machine learning model. This model then estimates brain network activities from simpler EEG measurements, avoiding the need for detailed brain shape, skull shape, or sensor position information.
2Measurement precision
If inverse problem analysis of EEG measurement results is performed, then brain network activity estimation is achieved, but large amount of operational processing is required, leading to large temporal granularity
Solution Approach 1:
The patent performs preliminary action by training a machine learning model using both EEG data and fMRI data before actual brain network activity estimation. During the training phase, the system learns the relationship between EEG signals and brain network activities from fMRI-validated data. Once trained, the model can rapidly estimate brain network activities from new EEG data without requiring complex real-time processing, thus achieving fine temporal granularity.
3Measurement precision
If fMRI measurement is used to directly observe brain network activities, then accurate brain network activity information is obtained, but large-sized apparatus is required and subject must remain in measurement apparatus, limiting possibility of wide spread use
Solution Approach 1:
The patent segments the measurement process into two distinct phases: (1) a calibration phase using fMRI to establish the relationship between EEG and brain network activities, and (2) a practical estimation phase using only portable EEG devices. The fMRI data is used to train a machine learning model that captures the mapping from EEG signals to brain network activities. After training, the system can be deployed with simple, portable EEG equipment, achieving both accuracy and portability.
4Adaptability or versatility
If EEG measurement is used for brain network activity estimation, then portability and mobility are improved, but inverse problem analysis requires detailed information and extensive processing, reducing utility
Solution Approach 1:
The patent creates a computational model (machine learning model) that copies the relationship between EEG signals and brain network activities as observed during fMRI measurements. This trained model serves as a surrogate that can rapidly estimate brain network activities from EEG data without requiring complex inverse problem analysis. The copying approach preserves the accuracy benefits of fMRI while achieving the portability and simplicity of EEG measurements.
Data Source
AI summary
A brain network activity estimation system includes means for obtaining brain wave measurement data and functional magnetic resonance imaging measurement data simultaneously measured from a subject, means for constructing a feature estimation model which receives the brain wave measurement data as input data and determining a parameter which defines the feature estimation model, means for calculating a feature value for each module based on an output value from each module that is calculated when the brain wave measurement data is provided as the input data, means for calculating an image feature value for each brain network based on the functional magnetic resonance imaging measurement data, and means for determining one or more modules which express activities of a specific brain network among a plurality of modules by evaluating correlation between the feature value for each module and the image feature value for each brain network.