Tomosynthesis Image Subtraction for Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tomosynthesis imaging faces challenges in generating high-quality subtraction images due to blurring caused by mechanical errors and body movement, which can lead to difficulties in detecting lesions like fine calcifications in breast cancer, especially when radiation sources are stationary, and artifacts from filtering processes in the FBP method.
Innovation Solution
A radiation image processing apparatus and method that obtain projection images at multiple radiation source positions with and without filtering, reconstruct tomographic images using processes excluding filtering, perform subtraction processes on corresponding cross-sectional planes, and apply filtering to the resulting subtraction images to reduce blurring and artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the FBP method with filtering is used for reconstruction, then blurring in images is reduced, but artifacts are introduced in the periphery of coordinate positions
Solution Approach 1:
The patent segments the image processing into two distinct stages: (1) artifact-free reconstruction using non-filtered back projection to generate initial tomographic images, and (2) selective filtering applied only to subtraction images after contrast enhancement. This segmentation prevents artifacts from propagating through the entire processing chain while still achieving blur reduction where needed.
Solution Approach 2:
The patent performs preliminary reconstruction without filtering to establish an artifact-free baseline. By conducting the reconstruction phase before any filtering operations, the system ensures that no artifacts are introduced during the critical image formation stage, allowing subsequent filtering to be applied safely to enhancement images.
2Manufacturing precision
If subtraction processes are performed on tomographic images reconstructed with filtering, then blurring is reduced, but artifacts from the filtering process are emphasized
Solution Approach 1:
The patent separates the subtraction process into two paths: one using non-filtered reconstructed images to avoid artifacts, and another using filtered images only for contrast enhancement. The final subtraction is performed on the artifact-free images, ensuring that enhancement does not compromise image purity.
Solution Approach 2:
The patent introduces an intermediary step where contrast-enhanced images are generated separately using filtered data, but the actual subtraction operation uses the original artifact-free reconstructed images. This intermediary approach allows the benefits of filtering to be explored without allowing artifacts to contaminate the final result.
3Loss of time
If the radiation source is moved continuously during imaging, then imaging time is reduced, but misalignment between projection images of different energies occurs
Solution Approach 1:
The patent performs preliminary alignment and registration of projection images from different radiation source positions before performing energy-based subtraction. By establishing geometric correspondence in advance, the system enables continuous motion imaging while maintaining precise alignment through pre-computed transformation matrices.
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors and adjusts for motion-induced misalignment by comparing projection images across multiple energy levels. Realignment transformations are applied based on detected deviations, ensuring that subtraction operations remain accurate despite continuous radiation source movement.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables the generation of high-quality tomographic subtraction images with reduced artifacts and blurring, facilitating the detection of breast cancer by emphasizing abnormal portions without introducing unnecessary components.
Implementation Method 1
radiation image obtaining apparatuses that employ radiation such as X rays and gamma rays
Implementation Method 2
obtain a plurality of first projection images by moving a radiation source relative to a detecting means and irradiating a subject with radiation
Implementation Method 3
A tomographic image is generated by reconstructing the projection images by an reconstruction method such as the shift and add method, the simple back projection method or the FBP method (Filtered Back Projection method)
Implementation Method 4
One type of subtraction process is temporal subtraction. In temporal subtraction, an image, in which blood vessels of a breast are not emphasized, is subtracted from an image, in which the blood vessels of the breast are emphasized
Implementation Method 5
The other type of subtraction process is energy subtraction. Energy subtraction utilizes the fact that contrast agents have different radiation absorption rates with respect to radiation having different energies
Implementation Method 6
the FBP method back projects a projection image after applying a filter thereon, to reduce blurring in images that flows in the direction of movement of the radiation source
Data Source
AI summary
First and second image obtaining units respectively obtain a plurality of first projection images and a plurality of second projection images by tomosynthesis imaging operations according to first and second imaging conditions. A reconstructing unit reconstructs the plurality of first and second projection images employing processes of a reconstruction process that includes a filtering process other than the filtering process, to generate a plurality of first tomographic images and a plurality of second tomographic images for each of a plurality of cross sectional planes within a subject. A subtraction processing unit generates tomographic subtraction images from the first and second tomographic images. A filtering processing unit administers filtering processes on the tomographic subtraction images, to generate processed tomographic subtraction images.