Medical Image Display Segmentation for Diagnostic Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical image retrieval systems face challenges in efficiently and accurately identifying the disease name of a medical image by displaying a large number of similar case images, which can lead to diagnostic inefficiencies due to the overwhelming number of images and limited selection options.
Innovation Solution
A control method for an information terminal that detects a region of interest in a target medical image, retrieves similar images, and displays them in order of decreasing similarity, with a disease name selection area to refine and classify images, allowing for efficient comparison and improved diagnostic accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a large number of similar case images are displayed to improve diagnostic accuracy, then the completeness of case coverage is improved, but the complexity of the display system and the difficulty of efficient comparison increase
Solution Approach 1:
The display screen is divided into multiple distinct areas: a first display area for the target medical image, a second display area for similar medical images arranged horizontally by similarity, and a third display area for disease name selection. This segmentation allows physicians to systematically compare images while maintaining organization, resolving the contradiction between comprehensive case coverage and display complexity.
Solution Approach 2:
The system transitions from displaying similar images in a single linear sequence to a two-dimensional organized layout with horizontal arrangement by similarity and vertical classification by disease name. This dimensional change enables efficient comparison of multiple images while maintaining systematic organization, addressing both completeness and complexity concerns.
2Productivity
If similar images are displayed in order of decreasing similarity to prioritize relevant cases, then the efficiency of identifying relevant cases is improved, but the ability to classify by disease name becomes more complex
Solution Approach 1:
The second display area is further divided into multiple sub-areas, with each sub-area corresponding to a selected disease name and containing similar images classified under that disease. This segmentation enables simultaneous presentation of similarity-based ordering and disease-based classification without increasing overall system complexity.
Solution Approach 2:
The display system dynamically reorganizes similar medical images into disease-specific sub-areas based on user selection of disease names from the third display area. This dynamic reconfiguration allows the system to adapt to different diagnostic needs, maintaining efficiency while enabling flexible classification.
3Measurement precision
If multiple disease names are selected to refine image classification, then the precision of disease identification is improved, but the number of sub-areas and overall display complexity increases
Solution Approach 1:
Each selected disease name generates a dedicated sub-area within the second display area, creating distinct classification zones. This segmentation allows precise disease identification through multiple selected disease names while maintaining organized separation of cases, preventing display chaos despite increased classification granularity.
Data Source
AI summary
A display control unit vertically divides a case display area in accordance with the number of disease names selected by a user to create a number of sub-areas equal to the number of disease names. Each of the sub-areas is vertically elongated so that thumbnail images of similar cases of the corresponding disease name are displayed so as to be aligned in a column. The display control unit displays, in each sub-area, thumbnail images of similar cases of the corresponding disease name so that the thumbnail images are aligned in a column in order of decreasing similarity to a search query image displayed in a layout area.