Cone Beam Artifact Correction in Gated CT Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cone beam artifacts occur in computed tomography (CT) systems due to mis-weighting of frequencies along the z-axis in short-scan reconstructions, leading to degraded image quality, particularly in scans where the X-ray source path does not cross all planes containing the object point, resulting in artifacts like bright shading, dark shading, and streaking.
Innovation Solution
A spatially variant 3D filter is employed to correct frequency mis-weighting in reconstructed volumetric image data, using a reconstructor configured to apply redundancy weighting and a combiner to add/subtract filtered data, effectively addressing the cone beam artifacts by accurately reproducing missing frequencies for each voxel.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a short-scan axial reconstruction is performed to improve temporal resolution and mitigate motion blur, then the scan time is reduced and temporal resolution is improved, but cone beam artifacts occur due to improper frequency weighting along the z-axis
Solution Approach 1:
The patent applies a spatially variant 3D filter that adapts the frequency weighting locally for each voxel based on its position in the volume. This allows proper weighting of higher frequencies along the z-axis in regions where they are needed, while maintaining appropriate weighting elsewhere, thereby eliminating cone beam artifacts without requiring a full 360-degree scan.
Solution Approach 2:
The patent modifies the frequency weighting parameters in the 3D Fourier domain by applying a spatially variant filter that changes the weighting of different frequency components based on their location. This parameter adjustment corrects the mis-weighting of higher z-axis frequencies that causes cone beam artifacts, enabling accurate reconstruction from short scans.
2Use of energy by moving object
If redundancy weighting is applied to improve dose utilization, then the energy efficiency is improved, but cone beam artifacts occur when the source path does not cross all planes containing object points
Solution Approach 1:
The spatially variant 3D filter applies different frequency weighting to different regions of the 3D Fourier domain based on the local geometry of the scan. This allows proper reconstruction accuracy in regions where the source path does not cross all planes, while maintaining efficient dose utilization in regions where sufficient data is available.
Solution Approach 2:
The patent introduces a spatially variant filtering step as an intermediary process between the weighted backprojection and final image formation. This intermediary filter corrects the frequency mis-weighting introduced by redundancy weighting, enabling both efficient dose utilization and accurate reconstruction.
3Manufacturing precision
If a conventional 2D filter is used to correct cone beam artifacts, then the image quality is improved, but additional errors are introduced due to the limitations of 2D filtering in 3D volumetric data
Solution Approach 1:
The patent transitions from 2D filtering to 3D spatially variant filtering in the volumetric domain. By operating in three dimensions rather than two, the filter can properly address the 3D frequency mis-weighting caused by cone beam geometry without introducing the artifacts and errors that arise from attempting to correct 3D problems with 2D methods.
Solution Approach 2:
The patent changes the filtering parameters from fixed 2D filter coefficients to spatially variant 3D filter coefficients that adapt to the local geometry. This parameter change enables accurate correction of cone beam artifacts throughout the entire volumetric dataset without introducing the errors that plague conventional 2D filtering approaches.
Data Source
AI summary
A system includes a reconstructor (314) configured to reconstruct cone beam projection data to generate cone beam artifact corrected short scan cone beam volumetric image data. A method includes reconstructing, with a reconstructor, cone beam projection data to generate cone beam artifact corrected short scan cone beam volumetric image data. A computer-readable storage medium storing computer executable instructions which when executed by a processor of a computer cause the processor to: reconstruct cone beam projection data to generate cone beam artifact corrected short scan cone beam volumetric image data.