Personalized Blood Vessel Map Construction Using Reference Image Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MRI technologies lack an efficient method to construct personalized blood vessel maps for Koreans, particularly in terms of age and region-specific averages, which hinders accurate diagnosis and prediction of vascular diseases.
Innovation Solution
A blood vessel map construction apparatus and method that receives a TOF image, extracts blood vessel stems and centerlines, detects branch points, searches for reference images based on location and curvature information, calculates similarity, and registers the images to create a personalized blood vessel map, enabling the prediction of vascular diseases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional MRI imaging is used to obtain blood vessel images, then general anatomical structures can be visualized, but personalized blood vessel maps specific to Korean population characteristics (age and region) cannot be constructed
Solution Approach 1:
The system pre-processes multiple TOF images from different subjects to construct reference blood vessel images that represent average Korean population characteristics for specific age groups and regions. This preliminary construction of population-specific references enables subsequent personalized mapping without requiring new imaging protocols
Solution Approach 2:
The system creates standardized reference blood vessel images that copy and represent the average vascular characteristics of Korean populations. These reference images serve as templates that can be compared against individual patient images to identify deviations and predict diseases
2Measurement precision
If detailed blood vessel analysis is performed to achieve accurate disease prediction, then diagnostic precision improves, but processing time and computational complexity increase
Solution Approach 1:
The system extracts only the essential centerline features and branch point information from complex 3D blood vessel images. By reducing the problem to one-dimensional centerline comparisons rather than full volumetric analysis, processing time is dramatically reduced while maintaining diagnostic accuracy
Solution Approach 2:
The system transforms complex 3D vascular structures into simplified parameter representations including centerline coordinates, branch point locations, and curvature measurements. This parameter transformation enables rapid comparison against reference images while preserving the critical geometric information needed for disease prediction
3Manufacturing precision
If complex image processing algorithms are used to extract blood vessel features, then extraction precision improves, but device complexity increases
Solution Approach 1:
The blood vessel extraction process is segmented into distinct functional modules: TOF image acquisition, blood vessel segmentation, centerline extraction, and feature detection. Each module performs a specific function with well-defined inputs and outputs, making the overall complex system manageable and maintainable while achieving high extraction accuracy
Data Source
AI summary
Disclosed herein a method of constructing a blood vessel map, the method including: receiving, by a blood vessel map construction apparatus, a TOF image of the object; extracting, by the blood vessel map construction apparatus, a blood vessel stem, a center line of a blood vessel, and features of the center line included in the TOF image, and detecting branch points of the blood vessel based on the extracted blood vessel stem, center line of the blood vessel, and features of the center line; searching, by the blood vessel map construction apparatus, a reference blood vessel image corresponding to the TOF image in consideration of location information or curvature information of the branch points; calculating a similarity between the TOF image and the reference blood vessel image; and registering the TOF image in the blood vessel map based on the similarity.