Medical Image Display Device for Accurate Tomographic Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical image display methods fail to accurately match and compare tomographic images taken at different times, leading to inaccuracies in observing the condition of a patient, as they often deform images, resulting in artifacts and an inability to observe the real shape of organs or regions of interest accurately.
Innovation Solution
A medical image display apparatus and method that uses a deformation amount calculation unit to identify corresponding positions in tomographic images without deforming them, allowing for precise matching and comparison of images taken at different times by using a second tomographic image identification unit to align images based on deformation amounts within a defined region of interest, and a display control unit to project and distinguish the region of interest on the second image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If non-rigid registration technique is used to match positions of images at different time points, then position matching accuracy is improved, but image deformation occurs causing artifacts and loss of real organ shape
Solution Approach 1:
The patent segments the image matching process into two independent parts: (1) use deformation amount to identify corresponding positions between images, and (2) display original images without deformation at those corresponding positions. This segmentation allows position matching to be accurate while preserving the original organ shapes in the displayed images.
Solution Approach 2:
The patent introduces deformation amount as an intermediary parameter that bridges the two images without deforming them. The deformation amount calculated from non-rigid registration is used to identify corresponding positions, but the actual displayed images remain undeformed, using the deformation amount only as a reference for alignment.
2Measurement precision
If images are deformed to match positions at different time points, then position matching is achieved, but processing complexity and computational cost increase
Solution Approach 1:
The patent extracts only the necessary information (deformation amount and corresponding positions) from the complex non-rigid registration process, while discarding the actual image deformation operation. This extraction approach maintains position matching accuracy while significantly reducing processing complexity by avoiding full image deformation and artifact correction.
3Ease of operation
If deformed images are displayed for comparison, then comparative image reading is enabled, but real shape of organs and regions of interest cannot be observed accurately
Solution Approach 1:
Instead of deforming images to achieve alignment and then trying to preserve shape information, the patent inverts the approach: it uses deformation amount to find corresponding positions, then displays the original undeformed images at those positions. This inversion allows both comparative reading and accurate shape observation to coexist.
Data Source
AI summary
[Aim] When two images have been obtained by imaging a same subject at different points in time, and comparative observation is performed on the two images by using tomographic images, the tomographic images at accurately corresponding positions are extracted and displayed for comparison. [Solution Means] First tomographic image information for identifying a first tomographic image included in a first image and first region-of-interest information for identifying a first region of interest on the first tomographic image are obtained. An image deformation amount of one of the first image and the second image for deforming the one of the first image and the second image to match the first image and the second image with each other is calculated. A slice image in the second image corresponding to the first tomographic image is identified, based on image deformation amounts at plural positions in the first region of interest and the first tomographic image information, as a second tomographic image, and a region of interest in the second image corresponding to the first region of interest is identified, as a second region of interest. The first tomographic image and the second tomographic image onto which the second region of interest is projected are displayed.