Multi-Energy CT Scatter Correction via Energy-Dependent Kernel Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-energy computed tomography systems face artifacts in reconstructed images due to spectral dependency of the scattering process, which are not adequately addressed by existing convolution error compensation methods.
Innovation Solution
A detection values correction apparatus and method that includes a scatter contribution providing unit and a combining unit to account for scatter contributions based on intensity, energy, and location, allowing for energy-dependent correction of detection values to reduce scattering artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If convolution error compensation using generated kernels is used to correct detection values, then scatter artifacts are reduced in single-energy imaging, but scatter artifacts remain in multi-energy reconstructed images due to spectral dependency of scattering
Solution Approach 1:
The patent applies parameter changes by transitioning from energy-independent scatter kernels to energy-dependent scatter kernels. The scatter correction is performed separately for each energy level, where the scatter contribution is calculated as a function of the detected energy. This allows the scatter correction parameters to adapt to the spectral characteristics of multi-energy imaging, resolving the contradiction between achieving accurate scatter correction and handling spectral dependency.
2Measurement precision
If energy-dependent scatter correction is implemented for multi-energy imaging, then scatter artifacts are reduced in reconstructed images, but calculation complexity increases due to separate processing for each energy level
Solution Approach 1:
The patent applies segmentation by dividing the scatter correction process into separate energy-level-specific calculations. Instead of performing a single scatter correction for all energies, the method segments the correction into multiple independent calculations, each tailored to a specific energy level. This segmentation enables accurate energy-dependent scatter correction while maintaining computational tractability by processing each energy level separately rather than requiring a complex unified approach.
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 solution effectively reduces scattering artifacts in reconstructed images by considering the spectral dependence of the scattering process, resulting in improved image quality.
Implementation Method 1
a radiation source for emitting radiation for traversing an object
Implementation Method 2
a detection unit comprising a detection surface for generating the detection values depending on the radiation after having traversed the object
Implementation Method 3
a scatter contribution providing unit for providing scatter contributions for different intensities, different energies and different locations on the detection surface
Data Source
AI summary
The present invention relates to a detection values correction apparatus for correcting detection values of a projection image of a multi-energy imaging system. A scatter contribution providing unit provides scatter contributions for different intensities, different energies and different locations on the detection surface of the detection values. A scatter contributions combining unit combines scatter contributions for correcting a detection value, wherein the combined scatter contributions represent the contribution of the scatter, which is caused by radiation of the other detection values of the projection image, to the detection value to be corrected and wherein the scatter contributions are combined under consideration of the intensity, energy and location on the detection surface of the other detection values. A correction unit scatter corrects the detection value of the projection image by using the combined scatter contributions.