Combined Sinogram and Image Domain Material Decomposition for Spectral CT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods of material decomposition in computed tomography (CT) either in the sinogram domain or the image domain fail to simultaneously account for beam-hardening effects and prior information about the imaged object, leading to inaccuracies and limitations in image reconstruction.
Innovation Solution
A combined method of sinogram- and image-domain material decomposition is performed, where beam-hardening-free sinograms are used to transition between domains, allowing for accurate beam-hardening corrections in the sinogram domain and incorporation of prior information in the image domain, thereby enhancing image quality and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If material decomposition is performed in the sinogram domain, then beam-hardening effects are accurately represented, but the ability to utilize a priori information about the imaged object is lost
Solution Approach 1:
The material decomposition process is segmented into two distinct stages: first in the sinogram domain to handle beam-hardening effects, then in the image domain to incorporate a priori information. This segmentation allows each domain to be optimized for its respective strength without compromise.
Solution Approach 2:
The patent introduces an intermediary step where material decomposition is first performed in the sinogram domain to correct beam-hardening effects, and then the results are used as input for image-domain decomposition that incorporates a priori information. This intermediary process enables the transfer of benefits between the two domains.
2Loss of information
If material decomposition is performed in the image domain, then a priori information can be utilized, but accuracy is reduced due to inability to account for beam hardening and scatter effects
Solution Approach 1:
Beam-hardening correction is performed as a preliminary action in the sinogram domain before image reconstruction and image-domain material decomposition. This preliminary correction removes the harmful effects that would otherwise degrade the accuracy of subsequent image-domain processing.
3Device complexity
If only single-domain material decomposition is performed, then the process is simpler, but both beam-hardening accuracy and a priori information utilization cannot be simultaneously achieved
Solution Approach 1:
The decomposition process is segmented into sinogram-domain and image-domain stages, each optimized for specific requirements. This segmentation resolves the contradiction by distributing complexity across stages rather than requiring a single complex unified approach.
Solution Approach 2:
The patent merges the advantages of both sinogram-domain and image-domain decomposition by combining them in a sequential two-stage process. The final result incorporates both beam-hardening accuracy from the sinogram domain and a priori information utilization from the image domain.
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 simultaneous realization of the advantages of both sinogram- and image-domain material decomposition, improving the accuracy and quality of reconstructed images by accurately handling beam-hardening and scatter corrections while incorporating prior information.
Implementation Method 1
A radiation source, such as an X-ray source, irradiates the body from one side. At least one detector on the opposite side of the body receives radiation transmitted through the body.
Implementation Method 2
The attenuation of the radiation that has passed through the body is measured by processing electrical signals received from the detector.
Implementation Method 3
The spectral signature of the respective materials is used to determine corresponding material projection lengths for each ray, wherein a predefined magnitude and spectral shape is used for the X-ray absorption coefficient.
Data Source
AI summary
A method and apparatus is provided to generate material-component images from spectral computed tomography (CT) projection data, using material decomposition in both the sinogram and image domains. From material components in the sinogram domain, monoenergetic sinograms are generated, and then monoenergetic images are reconstructed from the monoenergetic sinograms. Next, the monoenergetic images are decomposed into material-component images. Material decomposition in the sinogram domain enables beam-hardening corrections, and material decomposition in the image domain enhances image quality using prior information regarding the images including (e.g., smoothness and volume constraints) and using image-domain calibrations. Additionally, the method can be improved using scatter correction and detector-response and energy-spectrum calibrations. Further, iterations of the method can be performed by feeding back the material-component images to improve the scatter correction.