Medical Image Analysis System for Highlighting Lesion Differences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional CAD systems for medical image analysis often overwhelm physicians with numerous lesion candidates, leading to potential overlooking of important findings, as the analysis results are displayed after diagnosis completion, making it burdensome to efficiently identify critical information.
Innovation Solution
A system that compares first and second medical image analysis results from computer processing, highlighting differences in lesion candidates detected, such as new or changed lesions, through distinct output methods like pop-up windows, email notifications, or altered display cursors, to draw attention to significant changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all lesion candidates detected by CAD are presented to the physician, then the completeness of diagnosis is improved, but the burden on the physician increases and lesion candidates may be overlooked
Solution Approach 1:
The patent extracts only the newly detected lesion candidates from the CAD analysis results and presents them separately from previously detected lesions. This allows the physician to focus on new findings without being overwhelmed by all lesion candidates, maintaining diagnostic completeness while reducing burden.
Solution Approach 2:
The patent segments the CAD analysis results into two distinct groups: newly detected lesion candidates and previously detected lesions. By separating these categories and presenting them differently, the system helps the physician efficiently identify critical new findings while maintaining awareness of all lesions.
2Ease of operation
If lesion candidates are filtered by confidence or importance, then the number of presented candidates is reduced, but important lesions may be missed
Solution Approach 1:
The patent performs preliminary comparison between current and past CAD analysis results to identify newly detected lesion candidates before presenting them to the physician. This preliminary action ensures that all potentially important lesions are identified and presented, avoiding the risk of missing critical findings while still reducing the number of candidates displayed.
3Reliability
If analysis results are displayed after diagnosis completion, then fair diagnosis without CAD influence is achieved, but efficient identification of critical information becomes difficult
Solution Approach 1:
The patent extracts newly detected lesion candidates from the complete CAD analysis results and presents them in a prominent, separate manner after diagnosis completion. This allows the physician to maintain independent diagnosis without CAD influence while efficiently identifying critical new findings through the extracted and highlighted presentation.
Data Source
AI summary
A non-transitory computer-readable storage medium stores a program to cause a computer to perform: acquiring a first analysis result obtained by a computer process on a first medical image of a patient; acquiring a second analysis result obtained by a computer process on a second medical image of the patient; and comparing the first analysis result and the second analysis result and in a situation in which there is a difference, outputting in a manner different from the situation in which there is no difference.