Tomographic Reconstruction Algorithm Sensitivity Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tomographic phase contrast and dark-field imaging technologies face challenges with image artifacts, particularly cupping artifacts, despite efforts to account for magnification effects, and existing reconstruction methods are inefficient due to reliance on iterative algorithms that require simultaneous reconstruction of both phase contrast and dark-field imagery.
Innovation Solution
A system for tomographic reconstruction that weights phase contrast or dark-field projection data based on the sensitivity of measurements along rays, using a filtered-back-projection algorithm, allowing for separate reconstruction of phase contrast or dark-field imagery and incorporating a sensitivity compensation model to reduce artifacts, which can be applied in various imaging geometries including cone-beam, parallel, and helical configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative reconstruction algorithms are used for phase contrast and dark-field tomography, then reconstruction accuracy can be improved, but computational efficiency deteriorates and processing time increases
Solution Approach 1:
The patent replaces iterative reconstruction algorithms with a closed-form filtered back-projection algorithm, substituting a computationally intensive mechanical process with a more efficient mathematical approach. This allows accurate reconstruction of phase contrast and dark-field imagery without the computational burden of iteration, directly resolving the contradiction between accuracy and efficiency
Solution Approach 2:
The patent introduces sensitivity weighting as a new parameter in the reconstruction process, where projection data is weighted according to the sensitivity of measurements along each ray. This parameter adjustment enables the filtered back-projection algorithm to achieve accuracy comparable to iterative methods while maintaining computational efficiency
2Loss of information
If both phase contrast and dark-field imagery are reconstructed simultaneously using iterative algorithms, then complete diagnostic information is obtained, but device complexity and processing requirements increase
Solution Approach 1:
The patent separates the reconstruction process into independent channels for phase contrast and dark-field imagery. Each channel is processed separately using the filtered back-projection algorithm with appropriate sensitivity weighting, allowing complete diagnostic information to be obtained without the complexity of simultaneous iterative reconstruction of both modalities
Solution Approach 2:
The patent develops a universal filtered back-projection framework that can handle both phase contrast and dark-field reconstruction through a common algorithmic structure. The sensitivity weighting mechanism is adapted to each modality's specific requirements, providing a multi-functional solution that reduces overall system complexity while maintaining information completeness
3Manufacturing precision
If magnification effects are accounted for in reconstruction, then some artifacts are reduced, but cupping artifacts and sensitivity gradients persist
Solution Approach 1:
The patent incorporates sensitivity weighting that provides feedback compensation for measurement variations along different rays. By weighting projection data according to the sensitivity of each measurement, the algorithm actively corrects for cupping artifacts and sensitivity gradients, improving image quality while eliminating the residual artifacts that persist even when magnification effects are accounted for
Data Source
AI summary
A signal processing system (SPS) and related method for tomographic reconstruction of phase contrast or dark-field imagery. An improved weighting model is used to eliminate sensitivity gradient variations to so reduce artifacts is the reconstructed imagery. The system (SPS) is not reliant on iterative reconstruction algorithms. Instead, quicker reconstruction algorithms such as filtered back-projection can be used herein.