Vessel Cross-Section Measurement Using Multi-Ray Intensity Profiling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately measure the dimensions of narrow vessels non-invasively and effectively diagnose vessel dysfunctions, particularly in organs like the lung, due to limitations in image resolution and the inability to reliably assess vascular remodeling in conditions such as pulmonary hypertension.
Innovation Solution
A method involving 3D medical imaging to generate rays from vessel cross-sections, determine image intensity profiles, and apply a double inverse sigmoid function to calculate vessel wall thickness, lumen radius, and outer radius, enabling the creation of relationship maps to assess vascular health.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If 3D medical imaging is used to measure vessel dimensions, then non-invasive measurement is achieved, but measurement precision deteriorates for narrow vessels due to image resolution limitations
Solution Approach 1:
The patent divides the vessel measurement problem into multiple radial segments by generating multiple rays from the vessel center point in different directions. Each ray independently measures the intensity profile, and the final vessel dimensions are derived by combining measurements from all rays. This segmentation approach allows precise measurement of narrow vessels by averaging out noise and resolution limitations across multiple measurement paths.
Solution Approach 2:
The patent transitions from direct 2D cross-sectional measurement to a multi-dimensional approach by generating rays in multiple directions (adding angular dimension) and combining their measurements. This dimensional transformation enables more accurate determination of vessel boundaries and dimensions by utilizing information from multiple viewing angles within the same 3D image volume.
2Measurement precision
If multiple rays are generated to improve measurement accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent implements an automated measurement system where the computer automatically generates multiple rays, extracts intensity profiles, fits curves to determine vessel boundaries, and calculates final dimensions without manual intervention. The system self-performs the complex multi-step measurement process, reducing operational complexity despite the sophisticated measurement methodology. The automated workflow includes: generating rays from vessel center, extracting intensity profiles along each ray, fitting double inverse sigmoid functions to determine boundaries, and computing median values across all rays.
Solution Approach 2:
The patent transforms the measurement problem by changing parameters from direct pixel distance measurement to intensity-based curve fitting. By using double inverse sigmoid function fitting on intensity profiles and taking median values across multiple rays, the system converts a complex geometric measurement problem into a more robust parameter estimation problem that is less sensitive to image resolution limitations and noise.
3Productivity
If automated measurement is implemented, then productivity improves, but measurement precision may deteriorate due to algorithmic approximations
Solution Approach 1:
The patent employs an iterative curve fitting process where the double inverse sigmoid function is adjusted to match the observed intensity profile along each ray. The fitting algorithm continuously refines the parameter estimates (vessel boundary positions, wall thickness) by comparing model predictions with actual image data. This feedback mechanism ensures that automated measurements maintain high precision by continuously optimizing the fit between the mathematical model and the actual vessel structure in the image.
Solution Approach 2:
The patent performs preliminary processing steps including generating the vessel center point, creating multiple rays in predetermined directions, and extracting intensity profiles before the actual dimension measurement. These preliminary actions prepare the data in an optimized format that enables efficient and accurate automated measurement. The pre-computed ray paths and intensity profiles serve as the foundation for the subsequent curve fitting and dimension calculation steps.
Data Source
AI summary
A method for dimensional measurement of a vessel (100) of a subject that is a bodily fluid-conducting vessel comprising: receiving image data comprising a three-dimensional medical image containing the vessel; obtaining from the image data an image patch (130) that is a two-dimensional image containing a transverse cross-section of the vessel (100); generating from the image patch (130), a set of rays (R1 to RX), wherein each ray (R1 to RX) is a straight line, one end contacting a centre point (140) of the outer footprint (150) of the vessel (100) and at the other end extending beyond the outer footprint (150) of the vessel (100), wherein each ray (R1 to RX) has a different direction; determining for each ray (R1 to RX) an image intensity profile of the image patch (130) along the ray as a function of distance from the centre point (140); determining from the image intensity profile along each ray (R1 to RX), one or more of a ray wall thickness (RXwt), a ray lumen radius (RXlr), a ray outer radius (RXor); determining dimensional measurements of the vessel (100) comprising one or more of: a vessel wall thickness (Vwt) determined from a median of the ray wall thicknesses (RXwt) in the set of rays (R1 to RX), a vessel lumen radius (Vlr) determined from a median of the ray lumen radii (RXlr) in the set of rays (R1 to RX), a vessel outer radius (Vor) determined from a median of the ray outer radii (RXor) in the set of rays (R1 to RX).