Medical Image ROI Correction Using Image Features and User Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image diagnosis systems face challenges in accurately detecting regions of interest in medical images due to inclusion of extra peripheral ranges or omission of detection, requiring manual correction by users, which is time-consuming and inefficient.
Innovation Solution
An information processing apparatus that utilizes learning models to assist in correcting regions of interest by receiving user input corrections and specifying a second region of interest based on image features and correction instructions, reducing the need for manual adjustment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual correction of detected regions of interest is performed, then accuracy of region specification is improved, but time consumption and operational effort increase
Solution Approach 1:
The system performs preliminary detection of regions of interest using CAD before manual correction, automatically identifying candidate regions that require verification. This preliminary action reduces the overall correction time by pre-processing the image analysis and presenting only relevant regions for manual verification.
Solution Approach 2:
The system incorporates feedback mechanisms where the correction results from manual verification are fed back to improve the CAD detection algorithm. This feedback loop continuously refines the detection accuracy, reducing the need for extensive manual correction over time while maintaining high precision.
2Measurement precision
If manual correction of detected regions of interest is performed, then accuracy of region specification is improved, but operational effort increases
Solution Approach 1:
The CAD system performs self-service by automatically detecting and highlighting regions of interest that require verification. This automation reduces operational effort by eliminating the need for operators to manually search and identify all potential regions, focusing their effort only on verifying CAD-detected candidates.
Solution Approach 2:
The system replaces manual mechanical correction operations with automated CAD detection algorithms. By substituting the mechanical process of manual region drawing and adjustment with intelligent software-based detection, operational effort is significantly reduced while maintaining accuracy through subsequent verification.
3Productivity
If CAD detection is used to identify regions of interest, then detection speed is improved, but detection accuracy deteriorates due to extra peripheral ranges or omitted peripheral portions
Solution Approach 1:
CAD detection serves as a preliminary action that quickly identifies candidate regions of interest before final verification. The system uses CAD to pre-process and highlight potential regions, maintaining high detection speed while flagging areas that require subsequent manual verification to ensure accuracy.
Solution Approach 2:
The system implements feedback where manual verification results are used to refine and improve the CAD detection algorithm. This continuous feedback loop allows the system to maintain high detection speed while progressively improving detection accuracy by learning from correction patterns and reducing false positives and negatives.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
Provided are an information processing apparatus, an information processing method, and an information processing program capable of supporting interpretation of images. An information processing apparatus includes at least one processor, and the processor is configured to: acquire an image; display, on a display, a figure indicating a first region of interest included in the image in a superimposed manner on the image; receive a correction instruction for at least a part of the figure; and specify a second region of interest that at least partially overlaps with the first region of interest based on an image feature of the image and the correction instruction.