Deformable Airway Model Registration for Lung Deformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for registering 3D models of airways with the actual airways during minimally-invasive surgical procedures face challenges due to the flexibility and deformation of lungs, leading to reduced navigation accuracy.
Innovation Solution
The method involves receiving location data from a tool with a location sensor navigating the luminal network, identifying potential matches in a 3D model, and assigning a registration score based on a deformation model that accounts for rotation, compression, extension, and bending of airway regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If rigid registration is used to align 3D airway models with actual airways, then registration speed is improved, but navigation accuracy deteriorates due to lung deformation
Solution Approach 1:
The patent applies deformable registration that dynamically adapts the 3D airway model to match the actual deformed airway geometry during the procedure. The system continuously updates the model based on real-time measurements from the navigation tool, allowing the registration to account for lung deformation rather than relying on a fixed rigid alignment. This dynamic adaptation resolves the contradiction by maintaining high navigation accuracy despite changing anatomical conditions.
Solution Approach 2:
The system incorporates feedback loops where real-time location data from the navigation tool is used to iteratively refine the 3D airway model. The deformable registration algorithm continuously compares the model against actual measurements and adjusts the deformation parameters accordingly. This feedback mechanism enables the system to compensate for lung deformation and maintain accurate navigation throughout the procedure.
2Measurement precision
If deformable registration is applied to account for lung deformation, then navigation accuracy is improved, but registration time increases
Solution Approach 1:
The deformable registration process is segmented into multiple stages: initial rigid registration to establish a baseline alignment, followed by iterative deformation correction using real-time measurements. The system divides the complex deformable registration into manageable steps, performing coarse alignment first and then refining specific deformation parameters. This segmentation reduces the overall registration time while maintaining high accuracy by avoiding the need to compute the complete deformation field from scratch.
3Measurement precision
If real-time updates of potential matches are performed, then navigation accuracy is maintained, but computational load increases
Solution Approach 1:
The system performs partial deformable registration by focusing computational resources on specific regions of interest rather than the entire airway tree. The deformable registration algorithm is applied selectively to areas where the navigation tool is currently positioned or where deformation is most likely to occur. This partial action approach maintains location accuracy in the relevant regions while significantly reducing the overall computational energy required compared to performing deformable registration across the entire model.
Data Source
AI summary
A system for registering a luminal network to a 3D model of the luminal network includes a computing device configured to identify potential matches in the 3D model with location data of a location sensor, assigning one of the potential matches a registration score based on a deformation model applied to the 3D model, and displaying the potential match having the highest registration score.