3D Vessel Centerline Correction for Bends and Narrowing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for centerline extraction in 3D models of blood vessels, particularly coronary arteries, fail near bends, narrowing, or other dimensional changes, requiring significant manual correction due to sensitivity to anatomical differences and imaging artifacts.
Innovation Solution
A computer-implemented method involving iterative centering and equalizing operations to correct initial centerlines with sub-voxel precision, using a skeleton graph and image forces to enhance accuracy and reduce manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If existing centerline extraction methods are used, then the process can be automated, but the accuracy fails near bends, narrowing, or dimensional changes requiring manual correction
Solution Approach 1:
The patent applies a dynamic iterative correction process where the centerline is continuously refined through multiple passes of centering and equalizing operations. The algorithm dynamically adjusts the centerline position by computing distances to vessel walls and repositioning points to achieve sub-voxel precision, transforming a static extraction into a dynamic optimization process that adapts to local geometric variations.
Solution Approach 2:
The patent replaces manual mechanical correction with an automated computational system. Instead of requiring operators to manually adjust centerline points, the invention uses image forces and mathematical optimization to automatically compute and apply corrections, substituting human mechanical adjustment with algorithmic computation that operates at sub-voxel precision.
2Manufacturing precision
If manual correction is applied to improve centerline accuracy, then precision increases, but time consumption and complexity increase
Solution Approach 1:
The patent implements a self-correcting system where the centerline extraction algorithm automatically identifies and corrects its own errors without external intervention. The iterative process uses the vessel geometry itself to compute correction forces, allowing the system to self-optimize the centerline position through multiple passes of centering and equalizing operations until convergence is achieved.
Solution Approach 2:
The patent performs preliminary centering and equalizing operations in an iterative sequence before final output. By pre-computing correction forces and applying multiple passes of refinement, the system prepares the centerline in advance with sub-voxel precision, eliminating the need for subsequent manual correction and reducing total processing time.
3Measurement precision
If high resolution centerlines are computed, then accuracy improves, but computational complexity increases
Solution Approach 1:
The patent segments the centerline computation into distinct operational phases: initial extraction, iterative centering, and equalizing operations. Each phase handles a specific aspect of the computation, allowing the complex task of achieving sub-voxel precision to be broken down into manageable sequential steps that can be efficiently implemented and controlled.
Data Source
Figure 1~3
Figure 4~6
Figure 7a~7d
AI summary
A computer-implemented method comprising receiving (100) a patient-specific volume image, comprising structural information about a patient's vessels represented in voxels, including a use centerline to extract and/or present patient-specific data, according to the invention, includes a step of correcting (150), which comprises an iterative process including at least operation of centering (552) points defining the centerline, by locating them further from the closest surface defined in the geometry, and operation of equalizing points (553) defining the centerline, by bringing each point closer to mean position of adjacent points. The invention further concerns a computer program product and imaging system.