Tomographic Image Generation via Frequency Decomposition and Iterative Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tomosynthesis imaging faces challenges in generating high-quality tomographic images due to discrepancies in radiation density and artifacts caused by large structures overlapping with small structures, leading to difficulties in accurately calculating similarity between projection images and resulting in noise and obscured small structures.
Innovation Solution
A tomographic image generation device and method that perform frequency decomposition on projection images to generate band projection images, reconstruct these images for individual frequency bands, and then perform frequency synthesis to produce a tomographic image, while also adjusting emphasis levels and using regression analysis to improve image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If radiation dose is reduced to minimize radiation exposure, then radiation safety is improved, but quantum noise increases making images noisier
Solution Approach 1:
The patent segments the image reconstruction process into multiple iterations, where each iteration refines the tomographic image by comparing projected images with actual projection images. This iterative approach allows noise reduction while maintaining image quality even with low radiation doses.
Solution Approach 2:
The patent implements feedback mechanisms by calculating differences between projected images (generated from reconstructed tomographic images) and actual projection images, then using these differences to adjust and improve the reconstruction in subsequent iterations. This feedback loop enables effective noise suppression.
2Productivity
If back projection method is used to reconstruct tomographic images, then reconstruction speed is improved, but artifacts appear as virtual images of structures
Solution Approach 1:
The patent performs preliminary actions by generating initial tomographic images using back projection, then uses these images to create projected images that are compared with actual projection images. This preliminary reconstruction serves as a basis for subsequent iterative refinement, combining the speed of back projection with the accuracy of iterative methods.
Solution Approach 2:
The patent introduces projected images as an intermediary between the reconstructed tomographic images and the actual projection images. These projected images serve as a mediator to calculate differences and guide the iterative refinement process, reducing artifacts while maintaining reconstruction efficiency.
3Quantity of substance
If large structures are present in the subject, then structural information is captured, but small structures become obscured due to overlapping projections
Solution Approach 1:
The patent applies local quality by calculating differences between projected and actual projection images at each pixel location, allowing local refinement of image quality. This enables small structures to be enhanced locally without being obscured by large structures in other regions.
Solution Approach 2:
The patent performs multiple iterations of the reconstruction process, applying the refinement operation repeatedly until convergence or a predetermined number of iterations. This excessive action ensures that small structures are progressively enhanced and separated from large structures.
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 enhances image quality by separating small and large structure projections, reducing artifacts, and improving noise reduction, resulting in clearer and more accurate tomographic images.
Implementation Method 1
radiation, such as x-ray or γ-ray
Data Source
AI summary
An image obtaining unit obtains a plurality of projection images taken by imaging a subject with different radiation source positions. A frequency decomposition unit performs frequency decomposition on each of the projection images to obtain a plurality of band projection images representing frequency components for individual frequency bands for each of the projection images. A reconstruction unit reconstructs the band projection images to generate band tomographic images of the subject for the individual frequency bands. A frequency synthesis unit performs frequency synthesis on the band tomographic images to generate a tomographic image of the subject.