User-Steered Bronchoscopic Path Planning via 2D CT Control Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated path planning systems for bronchoscopic procedures require high-resolution CT scans and time-consuming bronchial tree segmentation, making them impractical for on-the-table diagnosis and often resulting in multiple or unreliable path candidates, which burdens physicians and is not feasible for procedures involving few ROIs.
Innovation Solution
A user-steered, on-the-fly path planning method that allows physicians to interactively define control points and extend paths using 2D sectional images from CT scans, employing cost analysis based on intensity and geometric characteristics, and graphic search applications to create feasible paths without the need for high-resolution imaging or segmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automated path planning systems use high-resolution CT scans and bronchial tree segmentation, then path planning accuracy is improved, but processing time and system complexity increase significantly
Solution Approach 1:
The patent extracts only the essential bronchial tree centerline information needed for path planning, rather than performing complete bronchial tree segmentation. This selective extraction approach maintains path planning accuracy while significantly reducing processing time and computational complexity.
Solution Approach 2:
The patent segments the path planning process into distinct phases: obtaining CT images, extracting bronchial tree centerlines, calculating cost functions, and generating path candidates. This segmentation allows for optimized processing at each stage, reducing overall processing time while maintaining accuracy.
2Reliability
If automated path planning systems perform complete bronchial tree segmentation, then path reliability is improved, but device complexity and processing burden increase
Solution Approach 1:
The patent extracts only the centerline information from CT images using intensity and gradient analysis, rather than performing complete bronchial tree segmentation. This approach maintains sufficient path reliability for clinical decision-making while significantly reducing system complexity and processing burden.
3Adaptability or versatility
If multiple path candidates are generated, then path selection flexibility is improved, but processing time and physician burden increase
Solution Approach 1:
The patent applies different cost function calculations to different regions of the bronchial tree based on local anatomical characteristics. This allows for efficient generation of multiple path candidates with varying degrees of optimality, providing physicians with flexible selection options while reducing overall processing time through localized optimization.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method, system, and program product are provided for user-steered, on-the fly path planning in an image-guided endoscopic procedure, comprising: presenting, on a display, a 2D sectional image showing a region of interest from a preoperative CT scan; defining a control point on the 2D sectional image within a patient's body lumen responsive to a first user input; centering the control point; adjusting a viewing angle about the control point to show a longitudinal section of the body lumen responsive to a second user input; identifying a second point on a planned path within the body lumen responsive to a third user input; extending a planned path connecting the control point and the second point; redefining the second point as a new control point; and repeating the presenting adjusting, identifying, extending, and the redefining steps until the planned path reaches a procedure starting point within the patient's body.