Region Correction Apparatus for Medical Tomographic Image Boundary Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for correcting the boundaries of regions of interest in three-dimensional medical images, such as those from CT or MRI scans, often result in over-extraction or under-extraction, leading to inefficient and operator-intensive correction processes, especially when dealing with tomographic images that are not smoothly connected.
Innovation Solution
A region correction apparatus and method that uses a display controller to show a first tomographic image, corrects its boundary using a correction instruction region, sets an instruction region on adjacent tomographic images based on the correction, and adjusts the size and shape of the instruction region to match the boundary of the region of interest, allowing for simple computation and reduced operator burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a spherical cursor is used to three-dimensionally correct the boundaries of regions of interest, then the boundaries can be corrected in three-dimensional space, but unintended overcorrection occurs because the boundaries in the direction in which tomographic images are arranged do not have a spherical shape
Solution Approach 1:
The patent applies local quality by using a planar cursor that adapts to the local geometry of the boundary being corrected. Instead of using a uniform spherical cursor throughout three-dimensional space, the cursor shape is locally optimized to match the planar nature of the boundary in the tomographic image arrangement direction, preventing overcorrection in regions where the boundary does not have spherical curvature.
Solution Approach 2:
The patent transitions from a three-dimensional spherical cursor to a two-dimensional planar cursor that operates within the context of the tomographic image slices. This dimensional adaptation allows the cursor to correctly handle boundaries that are planar in the arrangement direction while maintaining three-dimensional correction capability through the stacking of corrected two-dimensional regions.
2Extent of automation
If regions of interest are set in two tomographic images and interpolated to the tomographic images present between them, then regions of interest can be set in three-dimensional space, but this imposes a heavy burden on the operator
Solution Approach 1:
The patent applies self-service by enabling the operator to correct boundaries in only one tomographic image, and the system automatically propagates the correction to adjacent images through the instruction region mechanism. This reduces the operator's workload from setting regions in multiple images to correcting boundaries in a single image, while still achieving three-dimensional region correction.
3Manufacturing precision
If contour information is obtained for the plurality of tomographic images to perform deformation operation, then the contour line can be deformed across multiple planes, but this requires a long time for the processing
Solution Approach 1:
The patent extracts only the essential correction information from the operator's input in one tomographic image and uses this extracted instruction region to automatically generate corrections for adjacent images. This avoids the time-consuming process of obtaining and processing contour information for all tomographic images, while still achieving accurate contour deformation through the propagated instruction regions.
Data Source
AI summary
The display controller displays a first tomographic image of a three-dimensional image consisting of a plurality of tomographic images on a display unit. A first correction unit corrects the boundary of a first region of interest extracted from the first tomographic image, by a correction instruction using a correction instruction region for the boundary of the first region of interest. The first instruction region setting unit sets a first instruction region on a second tomographic image of the plurality of tomographic images. The second correction unit corrects the boundary of a second region of interest extracted from the second tomographic image by setting a second instruction region on the second tomographic image.