Hybrid CT Image Generation Combining Iterative and Filtered Back-Projection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computed tomography images reconstructed using iterative methods have a noise level of substantially zero, resulting in an undesirable 'painted' or 'flat' appearance, while images from filtered back-projection methods have higher noise, making it challenging to achieve a balanced noise appearance across different display window widths.
Innovation Solution
An image generation apparatus and method that combines images reconstructed using iterative and filtered back-projection methods, adjusting the weighting based on display window width to generate a combined image with a desired noise appearance, ensuring the image does not appear too noisy or flat across various window widths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative reconstruction method is used, then noise level is reduced to substantially zero, but image appears painted or flat with undesirable look-and-feel
Solution Approach 1:
The patent combines two different reconstruction methods (iterative and filtered back-projection) into a single hybrid approach. The iterative reconstruction provides noise reduction while the filtered back-projection component preserves natural texture and appearance, resolving the contradiction between noise reduction and visual quality.
Solution Approach 2:
The invention creates a composite image by merging the noise-reduced iterative reconstruction with the texture-rich filtered back-projection image. This composite approach allows both methods to contribute their strengths: noise reduction from iterative methods and natural appearance from filtered back-projection methods.
2Ease of manufacture
If filtered back-projection method is used, then image has natural appearance, but noise level is high making it difficult to achieve balanced noise appearance across different window widths
Solution Approach 1:
The patent merges filtered back-projection images (with natural appearance) and iterative reconstruction images (with low noise) to create a balanced result. The combination allows the image to maintain natural look-and-feel while achieving acceptable noise levels across various display window widths.
Solution Approach 2:
The invention adjusts the weighting parameters between the two reconstruction methods based on display window settings. By dynamically changing the contribution ratio of each method, the system optimizes noise appearance for different window widths while preserving natural appearance characteristics.
3Device complexity
If single reconstruction method is used, then processing is simple, but cannot provide balanced noise appearance across different display window widths
Solution Approach 1:
The patent implements a dynamic combination system where the weighting between iterative and filtered back-projection methods is adjusted based on display window parameters. This dynamic adaptation allows the system to optimize noise appearance for different window widths without requiring multiple fixed processing paths.
Solution Approach 2:
The hybrid reconstruction system serves multiple functions simultaneously: it provides noise reduction, maintains natural appearance, and adapts to different display window settings. A single combined approach replaces multiple separate processing methods, achieving versatility without proportionally increasing complexity.
Data Source
AI summary
The invention relates to an image generation apparatus for generating an image of an object. An image providing unit (11, 16) provides a first image of the object and a second image of the object, wherein the first image has a smaller noise level than the second image. A display window providing unit (12) provides a display window being indicative of the range of image values shown on a display (14), and a combining unit (13) generates a combination image by combining the first image and the second image depending on the window width of the provided display window. This allows considering the influence of the display window on the noise appearance. Thus, by taking into account the first image and the second image having different noise levels and the provided display window, a combined image can be generated, which has an improved noise appearance.