Blood Vessel Image Analysis for Early Atheroma Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical image processing technologies fail to effectively evaluate abnormalities in blood vessels that do not appear as constriction, such as those in the early stages of atheroma hardening, where lumen constriction may not be visible.
Innovation Solution
A medical image processing device that extracts blood vessel regions, evaluates shape and signal value distribution information, and detects abnormalities by comparing different blood vessel regions, using extended contour lines and reference shapes to identify abnormalities not appearing as constriction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If blood vessel analysis is performed only on lumen constriction, then the analysis method is simple, but abnormalities in early stages of atheroma hardening cannot be detected
Solution Approach 1:
The blood vessel region is segmented into multiple parts: lumen region, blood vessel wall region, and periphery region. Each region is analyzed separately using different evaluation methods, allowing detection of abnormalities at different stages of atheroma hardening while maintaining systematic analysis
Solution Approach 2:
The analysis is extended from one dimension (lumen constriction only) to multiple dimensions by evaluating shape characteristics and signal value distribution in different regions (lumen, wall, periphery). This multi-dimensional approach enables detection of early-stage abnormalities without excessive complexity
2Measurement precision
If only lumen constriction is evaluated, then the evaluation process is simple, but diagnostic precision for early-stage abnormalities is insufficient
Solution Approach 1:
The blood vessel region is pre-segmented into lumen, wall, and periphery regions before evaluation. Reference shapes are created in advance from normal sections, enabling rapid comparative evaluation of multiple regions without excessive processing time
Solution Approach 2:
Different evaluation parameters are applied to different regions: shape characteristics for lumen region, signal value distribution for wall and periphery regions. This parameter differentiation enables comprehensive diagnostic precision while optimizing processing efficiency
3Reliability
If comprehensive evaluation of blood vessel periphery is performed, then early-stage abnormalities are detected, but the complexity of analysis increases
Solution Approach 1:
Different evaluation methods are applied to different local regions: lumen region uses shape comparison, while wall and periphery regions use signal value distribution evaluation. This localized approach ensures high detection reliability for early-stage abnormalities without uniformly increasing complexity across all regions
Solution Approach 2:
Reference shapes and signal value distributions are created from normal blood vessel sections. These references serve as templates for comparing abnormal regions, simplifying the complexity of comprehensive evaluation while maintaining high detection reliability
Data Source
AI summary
Extraction means configured to extract a blood vessel region from medical image data, detection means configured to perform evaluation regarding the shape or signal value distribution information in the periphery of a blood vessel including blood vessel contour points and the margin of blood vessel contour points in the blood vessel region extracted by the extraction means and detecting an abnormal portion on the basis of the evaluation result, and display means configured to display information regarding the abnormal portion detected by the detection means are provided.