Iterative Reconstruction Weighting for CT Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional X-ray computed tomography (CT) iterative reconstruction algorithms suffer from artifacts and poor image quality due to strong count weighting, which is governed by a Poisson stochastic process.
Innovation Solution
The proposed solution involves calculating variance data, applying a low-pass filter and a range-compressing function to obtain statistical weight data, and using a threshold function to reduce artifacts in the reconstructed images, thereby modifying weight coefficients used in iterative reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional iterative reconstruction algorithms use strong count weighting based on Poisson stochastic process, then the reconstruction process is simplified, but the reconstructed images suffer from artifacts and poor image quality
Solution Approach 1:
The patent transforms the weighting parameter from direct count information to modified weight coefficients through a series of mathematical operations including logarithmic transformation, variance calculation, and thresholding. This parameter transformation resolves the contradiction by changing the nature of the weighting factor to achieve both artifact reduction and maintained algorithmic tractability
Solution Approach 2:
The patent introduces intermediate variables (variance data, statistical weight data) as mediators between the raw count information and the final weighting coefficients. These intermediaries process the count information through controlled transformations that preserve essential statistical properties while eliminating artifacts, thus improving image quality without excessive complexity increase
2Ease of operation
If direct count information is used as statistical weight in iterative reconstruction, then the weighting calculation is straightforward, but the reconstructed images contain artifacts
Solution Approach 1:
The patent converts the harmful strong weighting effect into a benefit by applying a threshold function that identifies and suppresses artifact-causing regions. The thresholding operation transforms the originally harmful strong weights into moderated weights where appropriate, while preserving genuine high-count regions, thus converting the artifact problem into a quality improvement mechanism
Solution Approach 2:
The patent performs preliminary transformations on the count information before using it as weight, including calculating variance data and applying statistical functions. This preliminary processing prepares the weight data in advance to avoid artifacts during reconstruction, maintaining ease of operation while eliminating harmful effects
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
An X-ray computed tomography apparatus (100) according to the embodiment is characterized by including an X-ray generation unit (105) generating X-rays, an X-ray detection unit (3) detecting X-rays generated from the X-ray generation unit (105) and transmitted through an object, a projection data generation unit (300) generating projection data based on an output from the X-ray detection unit (3), a variance data generation unit (400) generating variance data based on the projection data and noise data concerning predetermined noise, a statistical weight generation unit (500) generating a statistical weight by compressing the variance data to a predetermined data range, and a reconstruction unit (700) reconstructing volume data by iterative reconstruction processing based on the projection data and the statistical weight.