Spectral CT Material Decomposition With Guided Noise Regularization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Material decomposition in the projection domain of spectral CT is an ill-posed task, leading to noise amplification and streak artifacts, which degrade image quality and introduce regularization-induced bias.
Innovation Solution
A multi-step approach involving a first projection-domain material decomposition algorithm with Tikhonov regularization followed by a second algorithm that penalizes deviations from the initial estimates, using a guided Tikhonov regularization to reduce noise and outliers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If material decomposition is performed in the projection domain, then material-specific images can be generated, but noise amplification and streak artifacts occur
Solution Approach 1:
The patent applies preliminary action by performing denoising on projection data before material decomposition. A denoising filter is applied to reduce noise in the projection data, and then material decomposition is performed on the denoised data, preventing noise amplification from occurring in the first place rather than correcting it afterward.
Solution Approach 2:
The patent introduces an intermediary denoising step between data acquisition and material decomposition. This intermediary process acts as a mediator that reduces noise in the projection data before it enters the material decomposition algorithm, thereby preventing the noise from being amplified during decomposition.
2Reliability
If Tikhonov regularization is applied to reduce outliers, then extreme values are reduced, but regularization-induced bias is introduced
Solution Approach 1:
The patent applies local quality by using different regularization strategies for different parts of the solution process. Initial regularization is applied during iterative reconstruction to ensure convergence, then a second regularization step is applied specifically to the material decomposition results to correct bias while preserving accuracy. This localized approach to regularization at different stages resolves the contradiction between outlier reduction and quantitative accuracy.
Solution Approach 2:
The patent performs preliminary regularization during iterative reconstruction to establish a stable initial solution, then applies additional regularization specifically tailored to correct bias in the material decomposition results. This two-stage preliminary action approach ensures both outlier reduction and quantitative accuracy are achieved.
3Object-affected harmful factors
If noise reduction is applied to projection data, then image quality improves, but computational complexity increases
Solution Approach 1:
The patent applies preliminary denoising to projection data before material decomposition, which simplifies the subsequent decomposition process by reducing the noise that would otherwise need to be handled computationally intensive methods. This preliminary action reduces overall computational complexity while achieving noise reduction.
Solution Approach 2:
The patent segments the processing into distinct stages: denoising of projection data, iterative reconstruction with regularization, and material decomposition with bias correction. This segmentation allows each stage to be optimized independently, managing computational complexity while achieving noise reduction.
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
The method effectively reduces noise and regularization bias in material decomposition, improving image quality and accuracy without requiring additional computational resources.
Implementation Method 1
differentiation of materials in the scanned object... due to the photoelectric effect and due to Compton scattering
Implementation Method 2
differentiation of materials in the scanned object... due to the photoelectric effect and due to Compton scattering
Implementation Method 3
using different X-ray tubes, switching a tube between different peak energies (kVp)
Data Source
AI summary
The present invention relates to a method (1), resp. a device, system and computer-program product, for material decomposition of spectral imaging projection data. The method comprises receiving (2) projection data acquired by a spectral imaging system and reducing (3) noise in the projection data by combining corresponding spectral values for different projection rays to obtain noise-reduced projection data. The method comprises applying (6) a first projection-domain material decomposition algorithm to the noise-reduced projection data to obtain a first set of material path length estimates, and applying (7) a second projection-domain material decomposition algorithm to the projection data to obtain a second set of material path length estimates. The second projection-domain material decomposition algorithm comprises an optimization that penalizes a deviation between the second set of material path length estimates being optimized and the first set of material path length estimates.