Intraluminal Image Region Classification via Initial and Expansion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical diagnosis using capsule endoscopes is challenging due to the need to identify abnormal regions from a large number of intraluminal images, requiring significant user expertise and time, as existing image processing methods lack efficient automatic detection and classification of regions of interest.
Innovation Solution
An image processing device and method that classify regions of interest by detecting initial and expansion regions, calculating feature data, and determining their classification based on area size and color feature differences, prioritizing images for further examination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic detection and classification of regions of interest is implemented, then productivity and ease of operation are improved, but device complexity increases
Solution Approach 1:
The image processing device segments the region detection task into multiple stages: initial region detection using color tone features, expansion region detection using edge information, and classification determination. This segmentation allows each stage to focus on specific aspects of region identification, improving overall efficiency while keeping individual module complexity manageable
Solution Approach 2:
The system performs preliminary actions by first detecting initial regions based on color tone differences, then expanding these regions using edge information before final classification. This preliminary processing reduces the burden on the final classification stage and improves overall diagnosis efficiency
Solution Approach 3:
The patent introduces feature data calculation as an intermediary step between region detection and classification determination. By calculating color tone features and edge information as intermediate representations, the system simplifies the complex task of automatic classification while improving diagnostic accuracy
2Measurement precision
If manual identification of abnormal regions is performed, then measurement precision and reliability are maintained, but loss of time increases
Solution Approach 1:
The system incorporates feedback mechanisms where the classification determination unit uses the calculated feature data to verify and refine region identification. This feedback loop ensures that automatic detection maintains precision comparable to manual identification while significantly reducing the time required for diagnosis
Solution Approach 2:
The patent changes parameters such as color tone features and edge information thresholds to optimize the balance between detection speed and accuracy. By adjusting these parameters, the system maintains high measurement precision while reducing diagnosis time
Data Source
AI summary
An image processing device classifies a region of interest included in an image into multiple classification items. The image processing device has: an initial region detector that detects at least part of the region of interest and sets the part as an initial region; an expansion region detector that detects an expansion region by expanding the initial region; and a region determining unit that calculates feature data of the initial region and the expansion region, and determines, based on the feature data, to which of the multiple classification items the region of interest belongs.