3D Cortical Mesh Model for Brain Electrode Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for localizing subdural and depth electrodes in the brain suffer from significant errors due to non-linear brain surface deformation post-implantation, leading to inaccurate electrode placement, especially under the craniotomy flap, which can result in incorrect localization of electrodes on the cortical surface.
Innovation Solution
The use of a manipulable 3D cortical mesh model, which can be parcellated and morphed to accurately visualize and localize electrodes by accounting for anatomical features and deformations, allowing for precise registration of electrodes relative to the cortical surface, including buried and unburied regions, using techniques like recursive grid partitioning and geodesic projection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If co-registration of post-implant CT with pre-implant MRI is used to localize electrodes, then the localization process can be automated, but significant localization errors occur (up to 8mm maximum error, 4mm mean error) due to non-linear brain surface deformation
Solution Approach 1:
The patent creates a virtual copy of the patient's brain using a 3D mesh model generated from pre-implant MRI data. This virtual brain model serves as a template that can be deformed to match post-implant CT images, allowing automated localization while accounting for brain deformation. The virtual model is manipulated to conform to post-implant anatomical changes, enabling accurate electrode positioning without manual intervention.
Solution Approach 2:
The patent applies parameter changes by deforming the 3D mesh model to account for non-linear brain surface deformation caused by mass effect and fluid accumulation. The mesh model undergoes geometric transformation to match post-implant anatomical conditions, adjusting vertices and surface parameters to reflect actual brain changes. This enables accurate localization despite significant shape changes in the brain tissue.
2Measurement precision
If semi-automated techniques are used to project displaced electrodes onto high resolution MRI scans, then some localization improvement is achieved, but error remains substantial (up to 8mm maximum error) causing incorrect gyrus or lobe localization
Solution Approach 1:
The patent segments the brain surface into a 3D mesh model with discrete vertices and faces, allowing localized manipulation and deformation. The mesh structure enables independent adjustment of different brain regions to match post-implant conditions. This segmentation allows complex deformation to be broken down into manageable geometric transformations, improving localization accuracy while maintaining computational efficiency.
Solution Approach 2:
The patent transitions from 2D image processing to 3D volumetric modeling by creating a three-dimensional mesh representation of the brain. This adds a spatial dimension that captures depth and surface curvature information, enabling accurate representation of electrode positions on the cortical surface. The 3D mesh allows visualization and measurement in multiple dimensions simultaneously, improving localization precision beyond what 2D techniques can achieve.
3Shape
If electrodes are visualized on deformed brain surface, then anatomical accuracy is maintained, but electrodes appear incorrectly buried inside the brain rather than on the surface
Solution Approach 1:
Instead of deforming the electrode positions to match the deformed brain surface, the patent inverts the approach by deforming the brain surface model to match the electrode array configuration. The 3D mesh model is manipulated to conform to the actual electrode positions, effectively reversing the traditional co-registration problem. This inversion allows electrodes to be visualized correctly on the cortical surface while maintaining anatomical accuracy of the underlying brain structures.
Data Source
AI summary
This invention relates generally to methods for localization and visualization of implanted electrodes and penetrating probes in the brain in 3D space with consideration of functional brain anatomy. Particularly, this invention relates to precise and sophisticated methods of localizing and visualizing implanted electrodes to the cortical surface and/or topological volumes of a patient's brain using 3D modeling, and more particularly to methods of accurately mapping implanted electrodes to the cortical topology and/or associated topological volumes of a patient's brain, such as, for example, by utilizing recursive grid partitioning on a manipulable virtual replicate of a patient's brain. This invention further relates to methods of surgical intervention utilizing accurate cortical surface modeling and/or topological volume modeling of a patient's brain for targeted placement of electrodes and/or utilization thereof for surgical intervention in the placement of catheters or other probes into it.