MRI-Derived 3D Mesh Registration for Fluoroscopy Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surgical navigation systems relying on CT scans expose patients to radiation and face accuracy issues due to uneven MRI plane distribution, limiting their use in robotic guidance and navigation.
Innovation Solution
A method to generate a 3D mesh from MRI images by registering them with a nearest neighbor CT template, enabling seamless integration with fluoroscopy for radiation-free navigation and robotic guidance, using techniques like neural networks for registration seed determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CT scans are used for surgical navigation, then navigation accuracy is improved, but patient radiation exposure increases
Solution Approach 1:
The patent creates a virtual 3D copy of the patient's anatomy from MRI data that can be used for navigation without requiring repeated CT scans during surgery. This virtual model serves as a radiation-free alternative to actual CT imaging while maintaining navigational accuracy.
Solution Approach 2:
The patent replaces the mechanical/x-ray based CT imaging system with a software-based 3D modeling and registration system that uses MRI data. This substitution eliminates the need for ionizing radiation while providing equivalent or superior navigation capabilities through advanced image processing and neural network techniques.
2Object-affected harmful factors
If MRI is used for 3D scanning, then patient radiation exposure is reduced, but navigation accuracy deteriorates due to uneven plane distribution
Solution Approach 1:
The patent transforms the unevenly distributed 2D MRI planes into a uniformly distributed 3D mesh structure through coordinate transformation and interpolation algorithms. This parameter change converts the disadvantageous spatial distribution of MRI data into a format suitable for accurate 3D navigation.
Solution Approach 2:
The patent transitions from 2D MRI image planes to a 3D mesh representation by adding the third spatial dimension through mathematical modeling. This dimensionality change enables the system to compensate for uneven plane distribution and create an accurate three-dimensional model for navigation.
3Adaptability or versatility
If 2D-3D registration is performed, then integration of MRI with fluoroscopy is enabled, but complexity of the imaging system increases
Solution Approach 1:
The patent creates a universal 3D mesh model from MRI data that can be registered with multiple types of intraoperative imaging systems including fluoroscopy, C-arms, and other 2D imaging devices. This single 3D representation serves as a common reference frame for various imaging modalities, reducing overall system complexity.
Solution Approach 2:
The patent introduces a 3D mesh model as an intermediary representation between the preoperative MRI and intraoperative 2D imaging systems. This intermediary enables seamless registration and integration without requiring direct complex transformations between different imaging modalities.
4Measurement precision
If large imaging equipment like CT or MRI is used, then 3D scan quality is improved, but availability in operating rooms decreases
Solution Approach 1:
The patent performs the 3D scanning and model creation action beforehand in the radiology department using high-quality MRI equipment. The resulting 3D mesh model is then transported to the operating room, eliminating the need for large imaging equipment to be physically present during surgery while maintaining scan quality.
Solution Approach 2:
The patent creates a digital copy of the patient's anatomy in the form of a 3D mesh model that can be easily transported and used in the operating room. This digital copy replaces the need for physical presence of large imaging equipment, improving availability while maintaining the high scan quality achieved by dedicated radiology department equipment.
Data Source
AI summary
An imaging system can be configured to generate a three-dimensional (“3D”) mesh from a magnetic resonance imaging (“MRI”) image. The imaging system can include a computer platform configured to perform operations. The operations can include receiving the MRI image. The operations can further include obtaining a nearest neighbor computerized topography (“CT”) template relative to the MRI image. The operations can further include generating the 3D mesh of the anatomical feature based on the MRI image and the nearest neighbor CT template.


