Automated EEG Electrode Localization via 3D MRI Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for localizing electroencephalography (EEG) electrodes are crude, tedious, and do not accurately account for their relationship to the brain, leading to variability and inefficiency, especially in high-density arrays where image distortions and signal losses occur.
Innovation Solution
A fully automated method using three-dimensional MRI images to identify electrode locations, relying on object shapes and properties rather than intensity differences, allowing for precise localization of electrodes relative to the brain and overcoming artifacts caused by magnetic susceptibility and signal losses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual landmark-based methods are used for electrode localization, then the process is simple and quick, but the localization precision and accuracy relative to brain structures is poor
Solution Approach 1:
The patent introduces MRI imaging as an intermediary tool to capture both the head surface and embedded electrodes simultaneously. The MRI scanner serves as a mediator that provides a common reference frame for both anatomical structures and electrode positions, eliminating the need for separate digitization processes and improving localization accuracy without proportionally increasing system complexity
Solution Approach 2:
The patent creates a digital 3D model (copy) of the head and electrodes from MRI data. This virtual replica allows for precise measurement and analysis of electrode positions relative to brain structures without requiring physical manipulation or additional hardware during the analysis phase, thereby improving precision while controlling complexity
2Measurement precision
If digitizing wands or photogrammetry systems are used, then electrode locations are better localized relative to scalp features, but the methods are extremely tedious and time-consuming
Solution Approach 1:
The patent merges the anatomical imaging function with the electrode localization function into a single MRI scanning process. Both the head anatomy and the electrodes are captured in the same imaging session and processed together through automated algorithms, eliminating the sequential steps of separate digitization and reducing overall localization time while maintaining precision
Solution Approach 2:
The system performs self-localization by automatically detecting electrode positions from the MRI images through image processing algorithms. The computer system autonomously identifies electrode locations and calculates their coordinates without requiring manual pointing or user interaction, thereby dramatically reducing the time required while maintaining measurement accuracy
3Measurement precision
If MRI is used with exogenous markers (vitamin A or E capsules), then electrode locations can be determined, but image distortions and signal losses occur due to magnetic susceptibility variations from high electrode count and wiring
Solution Approach 1:
The patent extracts and removes the harmful exogenous markers (vitamin A or E capsules) from the system, relying instead on the electrodes themselves as the imaging targets. By eliminating these additional foreign materials that cause magnetic susceptibility artifacts, the method reduces image distortions and signal losses while still achieving accurate electrode localization through direct imaging of the electrode structures
Solution Approach 2:
The patent converts the potential harm caused by electrodes themselves into a benefit by using the electrodes as the primary imaging targets. Rather than trying to image around or mark separate from the electrodes, the method leverages the electrodes' presence and structure as the very objects to be localized, turning what could be a source of artifact into the foundation of the localization method
4Quantity of substance
If dense high-density electrode arrays (128+ electrodes) are used, then better brain coverage is achieved, but image distortions and signal losses increase due to magnetic susceptibility variations
Solution Approach 1:
The patent replaces the mechanical/digital pointing system with an optical/MRI imaging system. Instead of using physical digitizing wands or camera-based photogrammetry that require line-of-sight and manual operation, the method uses magnetic resonance imaging to directly visualize and locate all electrodes simultaneously, enabling high-density array processing without the cumulative errors and time requirements of point-by-point digitization
Data Source
AI summary
A method for fully automated localization and channel identification of electroencephalography (EEG) electrodes. The electrode locations are automatically identified from three dimensional images stored in an electronic format, wherein the images may be derived from magnetic resonance imaging (MRI) that render the electrodes visible and object shapes and properties are used to locate the electrodes in the three dimensional images. The three dimensional images also show the brain in detail, such that the relationship of the electrodes to the brain is available, thereby making it possible to better identify electrical sources within the brain that create the EEG signals.