Representative Medical Frame Selection Through Vessel-Aware Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for automatically analyzing medical images using machine learning models often select unsuitable frame images, leading to incorrect results due to the inability to generate representative frame images that accurately represent the medical images.
Innovation Solution
A method and electronic device that calculate scores for each frame image in a medical image using a machine learning model, considering similarity, image quality, and region size within the blood vessel, to generate a representative frame image by applying weight elements to merge these images, ensuring suitability for analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a frame image is selected from medical images for analysis, then the analysis process can be performed, but the selected frame image may be unsuitable leading to incorrect results
Solution Approach 1:
The patent applies preliminary action by pre-calculating scores for multiple frame images before the actual analysis. A scoring mechanism evaluates each frame image based on predefined criteria (such as contrast, clarity, and relevance to blood vessel detection) and ranks them in advance. This ensures that when analysis is needed, the most suitable frame image is already identified and selected, preventing selection of unsuitable images and ensuring reliable analysis results.
2Reliability
If manual selection of frame images is used, then suitability can be ensured, but the process is time-consuming and not scalable
Solution Approach 1:
The patent implements self-service by enabling the system to automatically evaluate and select frame images without human intervention. The scoring mechanism autonomously assesses each frame image based on objective criteria (contrast, clarity, blood vessel visibility) and selects the most suitable image automatically. This eliminates the need for manual selection while maintaining high reliability in frame image suitability, significantly improving processing speed and scalability.
3Measurement precision
If multiple frame images are evaluated using complex criteria, then the representative frame image can be accurately generated, but the computational complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the complex evaluation process into separate, independent scoring components. Instead of using a single complex evaluation function, the system segments the assessment into distinct criteria (contrast scoring, clarity scoring, blood vessel region scoring) that can be calculated independently. Each component evaluates specific aspects of the frame image, and the results are combined to produce the final score. This segmentation reduces computational complexity while maintaining high measurement precision in representative frame image generation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of generating a representative frame image of a medical image performed by at least one processor is disclosed, the method comprising: acquiring a medical image including blood vessels; calculating scores for each of a plurality of frame images included in the medical image; and generating a representative frame image of the medical image from the plurality of frame images based on the scores for each of the plurality of frame images.