Region Boundary Correction for Adjacent 3D Image Segments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately correct the boundaries of multiple adjacent regions of interest in three-dimensional medical images, especially when one region is reduced, leading to over- or under-extraction issues.
Innovation Solution
A region correction device and method that includes a processor to reduce a first region, derive a difference region, and assign adjacent regions to this difference region by expanding them sequentially in units of small regions, determining the boundaries of adjacent regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If one region of interest is corrected to be reduced, then the first region's boundary is improved, but the boundaries of adjacent regions become indeterminate
Solution Approach 1:
The patent segments the boundary correction process into distinct phases: first correcting the primary region's boundary, then separately correcting adjacent regions' boundaries. This segmentation allows each region's boundary to be determined independently, preventing the information loss that occurs when multiple regions share a common boundary adjustment.
Solution Approach 2:
The patent performs preliminary correction of the first region's boundary before addressing adjacent regions. By establishing the primary region's boundary first, the system creates a reference framework that guides subsequent boundary corrections of adjacent regions, ensuring that boundary information is preserved rather than lost.
2Productivity
If automatic extraction is performed, then productivity is improved, but extraction accuracy deteriorates due to over-extraction and under-extraction
Solution Approach 1:
The patent implements a feedback mechanism where automatically extracted regions are evaluated and manually corrected when over-extraction or under-extraction is detected. The correction results are then fed back into the system, allowing the automatic extraction algorithm to learn from errors and improve future extractions, thus maintaining both high productivity and accuracy.
Solution Approach 2:
The system performs self-correction by automatically identifying regions that require correction based on extraction quality metrics. The correction process is designed to be self-guided, where the system autonomously determines which regions need adjustment and applies corrections without requiring complete manual intervention, thereby maintaining productivity while improving accuracy.
Data Source
AI summary
A region correction device, method, and program make it possible to, when one of a plurality of regions is corrected to be reduced, determine the boundaries of a plurality of regions adjacent to the reduced region. A processor reduces a first region among a plurality of regions in response to an instruction to reduce the first region, the instruction being provided for a target image in which the plurality of regions are adjacent to each other, the plurality of regions being three or more regions different from each other. The processor derives a difference region representing a difference between the first region before reduction and the first region after reduction, the difference region being composed of a plurality of small regions. The processor assigns a plurality of adjacent regions adjacent to the first region to the difference region by sequentially expanding, in the difference region, the plurality of adjacent regions in units of the small regions from a boundary between the difference region and the plurality of adjacent regions.