Joint-Edge-Preserving Regularization for 4D CT Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional denoising methods for four-dimensional CT images underutilize common structures among three-dimensional constituents, leading to suboptimal noise reduction and potential blurring of edges in medical imaging applications.
Innovation Solution
The implementation of joint-edge-preserving (JEP) regularization in a penalized weighted least squares (PWLS) objective function, which utilizes a whitening transform to de-correlate representations and apply a JEP regularizer to improve denoising of sinograms and reconstructed images, preserving edge sharpness while reducing noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional denoising methods are applied to four-dimensional CT images, then noise reduction is achieved, but edge sharpness is lost and common structures among three-dimensional constituents are underutilized
Solution Approach 1:
The patent merges multiple three-dimensional CT images into a four-dimensional dataset, combining structural information across multiple time points or energy levels. By treating the data as a unified four-dimensional volume rather than separate three-dimensional slices, the method preserves common structures and edges across all constituents while denoising, thus resolving the contradiction between noise reduction and edge sharpness preservation.
Solution Approach 2:
The patent transitions from conventional three-dimensional denoising to four-dimensional denoising by adding a temporal or spectral dimension. This dimensional expansion allows the algorithm to exploit correlations across the fourth dimension, preserving edges and structures that would be blurred in traditional methods while achieving superior noise reduction through the additional structural information.
2Object-affected harmful factors
If conventional denoising methods are applied to four-dimensional CT images, then noise reduction is achieved, but structural redundancy among three-dimensional constituents is underutilized
Solution Approach 1:
The patent combines all three-dimensional constituents into a unified four-dimensional analysis framework, merging their structural information. This allows the algorithm to identify and preserve recurring structures across multiple constituents, fully utilizing the structural redundancy that would be wasted if each image were processed independently.
Solution Approach 2:
The four-dimensional denoising algorithm serves multiple functions simultaneously: it denoises each individual three-dimensional constituent while also preserving common structures across all constituents. This multi-functional approach ensures that structural redundancy is not lost but rather leveraged to improve overall image quality across the entire four-dimensional dataset.
Data Source
AI summary
A method and apparatus is provided to denoise computed tomography (CT) sinograms and reconstructed images by minimizing a penalized-weighted-least-squares objective function that includes a joint-edge-preserving regularization term. Common information shared among a series of related image and/or sinograms can beneficially be used to more accurately identify edges common to all of the image and/or sinograms. Thus, a joint edge preserving regularizer that relies on information from a series of three-dimensional constituents of a four-dimensional image, such as a time series of images in a CT profusion study or an energy series of images in spectral CT, can improve image quality by better preserving edges while simultaneously denoising the constituent images.