Medical Image ROI Sizing With Multiple Confidence Options
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems for medical images fail to provide alternatives when inference results regarding the size of a region of interest, such as a lesion region, are clearly wrong, lacking consideration for the possibility of errors in the inference process.
Innovation Solution
An image processing device that generates and displays multiple options of the size of a region of interest in a medical image, utilizing a feature extractor and binary classifiers to infer coordinates, and presents these options along with their confidence levels, allowing for error consideration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a regression model is used to infer the coordinates of a region of interest, then the measurement precision is improved, but the reliability deteriorates because only a single proposal is provided without alternatives
Solution Approach 1:
The patent segments the single inference result into multiple discrete size options (e.g., 5 possibilities ranging from 0.5cm to 2.0cm). By dividing the continuous inference output into segmented discrete choices, the system provides alternatives while maintaining measurement precision, resolving the contradiction between precision and reliability.
Solution Approach 2:
The patent changes the parameter representation from a single continuous coordinate value to multiple discrete size options. This parameter transformation allows the system to present alternatives (improving reliability) while maintaining the precision of the original regression model through structured option generation based on the inferred coordinates.
2Reliability
If multiple options of region of interest size are generated and displayed, then the reliability is improved by presenting alternatives, but the device complexity increases
Solution Approach 1:
The patent implements preliminary action by pre-defining discrete size options (0.5cm, 1.0cm, 1.5cm, 2.0cm) before the inference process. This preliminary preparation simplifies the actual inference processing, as the system only needs to map the regression output to predefined options rather than generating and evaluating multiple complex hypotheses, thus reducing overall device complexity while maintaining reliability.
Solution Approach 2:
The patent uses copying by creating simplified discrete representations (copies) of the continuous inference result. Instead of working with complex continuous coordinate spaces for multiple options, the system copies the inference outcome into predefined discrete size categories, significantly reducing computational complexity while still providing multiple alternatives to the user.
Data Source
AI summary
The image processing device 1X includes an image acquisition means 30X, an option generation means 33X, and a display control means 34X. The image acquisition means 30X acquires a medical image. The option generation means 33X generates, based on the medical image, plural options of a size of a region of interest included in the medical image. The display control means 34X causes a display device to display information on the plural options of the size of the region of interest. With the displayed information, healthcare worker's decision making is assisted, for example.