Working Memory MEG Classification Using Integrated Source Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for processing magnetoencephalography (MEG) data lack an integrated pipeline from preprocessing to source reconstruction, making the process cumbersome and not fully automatic.
Innovation Solution
A system that includes a magnetoencephalography data acquisition module, a preprocessing module, a source reconstruction module, and a machine learning classification module, enabling comprehensive processing and classification of MEG data for working memory tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional lower-level processing scripts are used for MEG data processing, then each script can perform a specific processing step, but the overall process becomes cumbersome and requires operator experience, reducing automation
Solution Approach 1:
The patent combines multiple separate lower-level processing scripts into a single integrated MEG data processing pipeline that performs preprocessing, source reconstruction, and classification in one unified system, eliminating the need for operators to manually chain multiple scripts together
Solution Approach 2:
The processing pipeline is designed to perform multiple functions (preprocessing, source reconstruction, classification) within a single system, making it universally applicable to different MEG data processing needs without requiring separate specialized scripts for each task
2Measurement precision
If MEG source reconstruction is performed to calculate internal brain signals, then spatial resolution is improved, but the inverse problem creates uncertainty due to infinite solutions
Solution Approach 1:
The patent introduces an integrated processing pipeline that acts as an intermediary framework, coordinating multiple processing steps including preprocessing and source reconstruction algorithms to systematically manage the inverse problem and reduce uncertainty through structured processing
Solution Approach 2:
The pipeline performs preliminary preprocessing steps (such as noise filtering and artifact removal) before source reconstruction, preparing the data in advance to constrain the inverse problem and reduce the number of infinite solutions, thereby improving reliability
3Measurement precision
If fMRI is used to study brain activity, then spatial resolution is high, but temporal resolution is low at 2 seconds
Solution Approach 1:
The patent uses MEG technology which copies the brain's magnetic field signals directly without the temporal averaging required by fMRI, preserving the fast temporal dynamics while maintaining spatial information through source reconstruction capabilities
4Speed
If EEG is used to capture brain changes, then temporal resolution is high, but spatial resolution is poor and source reconstruction is difficult
Solution Approach 1:
The patent replaces EEG's electrical field measurements with MEG's magnetic field measurements, which are less distorted by the conductive properties of brain tissue, skull, and skin, thereby improving spatial resolution while maintaining the high temporal resolution characteristic of electrophysiological methods
Data Source
AI summary
A system for classifying working memory task magnetoencephalography based on machine learning, including: the magnetoencephalography data acquisition module configured to acquire magnetoencephalography data of a subject in different working memory task states; the magnetoencephalography data preprocessing module configured to control the quality of magnetoencephalography data in different working memory tasks and separate noises and artifacts; the magnetoencephalography source reconstruction module configured for sensor signal analysis and source reconstruction analysis for the data processed by the magnetoencephalography data preprocessing module; and the machine learning classification module is configured to classify the working memory tasks to which the subjects belong by taking power time series as features. The present disclosure integrates the complete analysis pipeline from preprocessing to source reconstruction of the working memory magnetoencephalography data, classifies the working memory task magnetoencephalography data, and is of great significance to the study of working memory decoding and brain memory related mechanisms.