Endoscope Image List Linking Instruments to Regions of Interest
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing diagnosis systems fail to effectively present the relationship between medical equipment and regions of interest, leading to increased user workload and potential errors in report creation.
Innovation Solution
A diagnosis support system that utilizes image recognition to detect medical equipment and regions of interest, assigning detection information based on similarity and feature quantities, and displays these relationships in a list for efficient diagnosis and reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If image recognition is used to detect medicine and equipment from endoscopic images, then automatic input assistance is provided, but the relationship between detected objects and regions of interest cannot be grasped
Solution Approach 1:
The system merges the detection results of medicine/equipment with the detection results of regions of interest by associating them through spatial relationship analysis. The information processing unit combines both detection outcomes and determines spatial relationships between them, creating a unified view that preserves the connection between treated objects and target regions.
Solution Approach 2:
The information processing unit acts as an intermediary that receives detection information from both the medicine/equipment detector and the region of interest detector. It processes this information to determine spatial relationships and generates integrated output that maintains the connection between previously unrelated detection results.
2Loss of information
If manual association of medicine/equipment with regions of interest is performed, then relationship information is preserved, but user workload increases
Solution Approach 1:
The system performs self-service by automatically determining spatial relationships between detected medicine/equipment and regions of interest without requiring manual user intervention. The information processing unit autonomously analyzes detection results and establishes associations, eliminating the need for users to manually perform this tedious task while preserving relationship information.
Solution Approach 2:
The system replaces the manual mechanical process of associating medicine/equipment with regions of interest with an automated information processing system. The processor automatically analyzes spatial relationships and generates associations, substituting human cognitive and manual effort with computational analysis.
3Loss of information
If all detected images are displayed in a list, then complete information is provided, but useful information cannot be easily grasped
Solution Approach 1:
The system applies local quality by selectively emphasizing images that contain both medicine/equipment and regions of interest in the displayed list. Rather than treating all images uniformly, it highlights relevant images with special markers or positioning, making useful information stand out while maintaining access to complete information through the full list.
Solution Approach 2:
The system segments the image list display into different categories or priority levels. Images containing both medicine/equipment and regions of interest are separated or marked distinctly from other images, allowing users to quickly identify relevant information while the complete set remains available for comprehensive review.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An object of the invention is to provide a diagnosis support system, an endoscope system, a processor, and a diagnosis support method which can present useful information to a user. With the diagnosis support system according to an aspect of the invention, since first detection information indicating a detected medicine and/or equipment and second detection information indicating a detected region of interest are assigned to an image for medical use, and images are displayed in a list in an aspect according to the detection information, it is possible for a user to easily grasp the detection information (first detection information and second detection information) and a relationship between the detection information and the image. In this manner, according to the first aspect, it is possible to present useful information (detection information) to a user, and it is possible for a user to efficiently perform diagnosis, report creation, and the like.