Endoscopic Image Selection via Priority Orders
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-aided diagnosis systems fail to efficiently support report creation in endoscopic examinations when they do not detect lesions, leaving doctors with a burdensome task of selecting appropriate images for reports.
Innovation Solution
A medical support system and method that acquire endoscopic images and additional information, allowing for automatic selection of report attachment images based on priority orders set for the information when no lesion is detected, facilitating efficient report creation by prioritizing organ, site, and imaging timing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a CAD system is used to automatically detect lesions in endoscopic images, then the productivity of report creation is improved when lesions are present, but the system fails to provide efficient support when no lesions are detected
Solution Approach 1:
The system dynamically adapts its image selection strategy based on CAD detection results. When lesions are detected, it uses CAD-based selection; when no lesions are detected, it switches to additional information-based selection. This dynamic adaptation ensures continuous productivity improvement across different examination scenarios
Solution Approach 2:
The system changes the selection criteria parameter from CAD-based lesion detection to additional information-based priority ordering depending on the examination outcome. This parameter change allows the system to maintain versatility and provide efficient support whether or not lesions are present
2Reliability
If doctors manually select endoscopic images for reports from multiple captured images, then the reliability of image selection is improved, but the time required for report creation increases
Solution Approach 1:
The system performs preliminary organization of endoscopic images by acquiring additional information (organ, site, imaging timing) and establishing priority orders before the doctor needs to select images. This preliminary action reduces the doctor's workload during actual image selection while maintaining reliability
Solution Approach 2:
The system acts as an intermediary between the captured endoscopic images and the doctor's selection process. It uses additional information as a mediator to automatically prioritize images, presenting a pre-organized selection to the doctor that maintains accuracy while reducing time consumption
3Quantity of substance
If multiple endoscopic images are captured during an examination, then the completeness of examination coverage is improved, but the complexity of selecting appropriate images for reports increases
Solution Approach 1:
The system segments the large set of captured endoscopic images into organized groups based on additional information (organ, site, imaging timing). This segmentation reduces the complexity of selection by presenting images in structured categories with established priority orders, making it easier for doctors to navigate and select appropriate images
Data Source
AI summary
An image acquisition unit acquires a plurality of endoscopic images captured in an endoscopic examination. An additional information acquisition unit acquires respective pieces of additional information of the plurality of endoscopic images. An image selection unit selects a predetermined amount of report attachment images from among the plurality of endoscopic images based on priority order set for the pieces of additional information when no endoscopic image including a lesion that is to be attached to a report is detected by the image analysis device in the endoscopic examination.