Coronary Angiography Vessel Classification Using Imaging Angle
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for diagnosing cardiovascular diseases through coronary angiography images lack accuracy due to the diversity of vascular types and structures, necessitating improved vessel classification techniques.
Innovation Solution
A method and electronic device that utilize angle information of the image acquisition device relative to the subject, combined with machine learning models, to identify and classify cardiovascular vessels accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional physician experience-based diagnosis is used, then diagnostic capability can be achieved, but accuracy is limited due to subjectivity and inconsistency
Solution Approach 1:
The patent replaces the mechanical system of human physician judgment with an electronic computing system that processes CAG images through machine learning algorithms. The processor automatically classifies vessels and identifies abnormalities, substituting human subjective analysis with objective computational analysis, thereby improving both accuracy and consistency of diagnoses
Solution Approach 2:
The system enables self-service diagnosis through automated machine learning models that independently analyze CAG images without requiring continuous human intervention. The electronic device performs vessel classification and abnormality detection autonomously, allowing the diagnostic system to serve itself while maintaining high reliability and consistency across multiple diagnoses
2Productivity
If machine learning analysis is implemented, then diagnostic speed is improved, but accuracy deteriorates due to diversity of vascular types and structures
Solution Approach 1:
The patent segments the complex task of cardiovascular diagnosis into distinct classification categories (RCA, LCA, LMCA, LAD, LCX) based on vessel type and imaging angle. By dividing the diverse vascular structures into manageable segments with specific classification criteria, the machine learning model can accurately identify and classify each vessel type while maintaining high diagnostic speed
Solution Approach 2:
The system changes parameters by incorporating angle information of the image acquisition device as a key feature for vessel classification. By adjusting and utilizing multiple parameters (vessel morphology, imaging angle, contrast enhancement patterns), the machine learning model achieves accurate classification of diverse vascular structures while maintaining rapid diagnosis speed
3Reliability
If comprehensive vessel analysis is performed, then diagnostic completeness is improved, but system complexity increases
Solution Approach 1:
The patent implements a universal machine learning model that performs multiple diagnostic functions simultaneously: vessel classification, abnormality detection, and diagnosis generation. This multi-functional system achieves comprehensive diagnostic completeness without requiring separate complex systems for each function, thereby managing system complexity while maintaining thorough analysis
Data Source
AI summary
A method of classifying vessels, performed by at least one processor, the method including: acquiring an image including the cardiovascular vessels of a subject and angle information of an image acquisition device for the subject; and based on the angle information, identifying at least one type of cardiovascular vessel included in the image is disclosed.