Tubular Structure Tracking Using Trained Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vessel tracking and segmentation methods often require manual correction due to difficulties in accurately navigating loops in tubular structures, leading to incorrect path shortcuts.
Innovation Solution
A method and apparatus using trained models to determine the path of tubular structures in medical imaging data by obtaining initial positions, selecting sub-regions, and updating positions based on curvature and torsion parameters, allowing for fully automated tracking without manual correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional vessel tracking algorithms are used, then the tracking process can be performed, but manual correction is required due to incorrect path shortcuts in loops
Solution Approach 1:
The system uses feedback mechanisms where the tracking algorithm continuously monitors the vessel path and compares it against anatomical constraints. When a loop is detected or the path deviates from expected anatomical structures, the system provides feedback to correct the trajectory, preventing incorrect shortcuts and ensuring accurate tracking through loops without manual intervention.
Solution Approach 2:
The system performs preliminary actions by pre-processing the imaging data to identify anatomical landmarks, vessel caliber changes, and potential loop structures before the actual tracking begins. This preliminary analysis allows the algorithm to anticipate correct paths through loops and avoid incorrect shortcuts, enabling fully automated accurate tracking.
2Measurement precision
If manual correction is performed, then tracking accuracy can be improved, but time consumption increases
Solution Approach 1:
The tracking system performs self-service by automatically detecting and correcting its own errors. The algorithm independently identifies when a shortcut has occurred or when the path enters an anatomically impossible region, and self-corrects by re-tracing the path through the loop using learned anatomical patterns, eliminating the need for manual correction while maintaining high precision.
Solution Approach 2:
The system implements feedback loops that continuously monitor tracking accuracy against anatomical constraints. When deviations are detected, the system automatically adjusts the path without external intervention, providing real-time correction that maintains precision while eliminating time-consuming manual review and correction processes.
3Speed
If simple tracking algorithms are used, then computational speed is maintained, but accuracy deteriorates in loop regions
Solution Approach 1:
The system applies local quality by using different tracking strategies for different regions. In straightforward vessel segments, simple tracking algorithms maintain high speed. In loop regions and anatomically complex areas, the system automatically switches to enhanced tracking modes that incorporate loop detection and correction mechanisms, ensuring high accuracy where needed without sacrificing overall computational speed.
Solution Approach 2:
The tracking algorithm dynamically adjusts its complexity based on the local anatomical context. The system monitors vessel geometry in real-time and adapts the tracking approach - using simple methods when the vessel path is straightforward and activating sophisticated loop-handling techniques only when loops or complex geometries are detected, maintaining computational efficiency while ensuring accuracy in challenging regions.
Data Source
AI summary
A method of determining a path of a tubular structure comprises: obtaining volumetric medical imaging data that represents anatomy of a subject including the tubular structure; performing a position-determining procedure that comprises: obtaining an initial position of a current point on said path of the tubular structure; selecting a sub-region based on said initial position; obtaining values of at least one parameter representative of the tubular structure; inputting to a trained model both data from said selected sub-region and said parameter values, the trained model having been trained to determine paths of tubular structures; outputting by the trained model a position for a next point on said path of the tubular structure; and updating said initial position to said next point; and repeating the position-determining procedure for updating the initial position thereby to obtain the path of the tubular structure.