High Density Forward Projector for CT Spatial Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computed tomography (CT) systems face limitations in spatial resolution due to detector pixel size and focal spot size, leading to resolution degradation in zoomed reconstructions, such as those used in imaging sinuses, coronary arteries, or cochlear implants, where voxel size becomes small and detector sampling pitch fundamentally limits spatial resolution.
Innovation Solution
The implementation involves an initial high-resolution reconstruction, followed by high-density forward projection and sinogram updating using both original and generated data to achieve a high-resolution sinogram, which is then used for a final reconstruction, employing Fast Fourier Transform (FFT) convolutions and back-projections to enhance spatial resolution without introducing aliasing artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If detector pixel size is reduced to improve spatial resolution, then spatial resolution improves, but detector sampling pitch fundamentally limits spatial resolution and increases system complexity
Solution Approach 1:
The patent applies dimensionality change by transitioning from detector-space sampling to image-space reconstruction. Instead of relying solely on detector pixel size (spatial dimension), the system uses iterative reconstruction algorithms that operate in the image domain, effectively adding a computational dimension that overcomes the physical sampling limitations of the detector.
Solution Approach 2:
The patent changes the fundamental parameter of image formation from direct detector sampling to iterative reconstruction. By using algorithms that repeatedly forward-project and back-project data through the object, the system transforms the limited detector sampling into high-resolution images through mathematical operations rather than physical sampling density.
2Measurement precision
If zoomed reconstruction is used to image small structures, then spatial resolution improves for specific regions, but voxel size becomes small and resolution degradation occurs due to detector pixel size limitations
Solution Approach 1:
The patent segments the reconstruction process into multiple iterative steps, where each step refines the image quality. The segmentation of reconstruction into forward projection, back projection, and iterative refinement allows independent optimization of spatial resolution without being constrained by fixed detector pixel dimensions.
Solution Approach 2:
The patent replaces the mechanical limitation of detector pixel size with a computational system. Instead of relying on physical detector resolution, the system uses iterative algorithms that mathematically reconstruct high-resolution images from limited detector data, substituting mechanical sampling with computational processing.
3Measurement precision
If conventional deconvolution approach is used to mitigate focal spot size and detector pixel size, then spatial resolution improves, but aliasing artifacts are introduced
Solution Approach 1:
The patent implements feedback through iterative reconstruction, where each iteration uses the previous result to generate improved data. The forward projection step continuously refines the sinogram by comparing generated and original data, creating a feedback loop that progressively improves resolution while suppressing aliasing artifacts through controlled iterative refinement.
Data Source
AI summary
A medical imaging apparatus, processing device or specialized circuit can include an input interface to input scan data of a medical image scan of a target object, a processor to generate an output image from the input scan data, and an output interface to output the output image to, e.g., a display. The processor can execute a first reconstruction of the scan data to obtain an intermediate image of the target object, a high-density forward projection of the intermediate object to obtain generated data, a sinogram updating using both of the generated data and the scan data to obtain a high-resolution sinogram, and a second reconstruction based on the high-resolution sinogram to obtain an output image.