Anchored Kernel Scatter Estimate for CT Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radiological imaging techniques, such as CT scans, face challenges in reducing noise and artifacts caused by scatter photons, which degrade image quality and accuracy, particularly in cone-beam CT systems without anti-scatter grids, leading to increased noise amplification and resolution loss during scatter correction.
Innovation Solution
The proposed solution involves a radiological imaging apparatus with a data processing system that separates radiation data into primary and scatter components, using generating functions and scatter models like convolutional neural nets to estimate and subtract the scatter component, thereby generating a scatter-corrected image, while treating the scatter-only component separately for enhanced noise reduction and resolution preservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If scatter correction is applied to reduce noise and artifacts, then image quality is improved, but resolution is degraded
Solution Approach 1:
The patent segments the scatter correction process into two distinct components: a non-scatter-corrected component and a scatter-only component. The scatter-only component is reconstructed from scatter-corrected projection data, while the non-scatter-corrected component is reconstructed from the original projection data. These two components are then combined to produce the final image. This segmentation allows different processing strategies to be applied to different components, preserving resolution in the non-scatter-corrected component while removing noise and artifacts in the scatter-only component.
2Object-affected harmful factors
If scatter correction is applied to reduce noise, then noise is reduced, but resolution loss increases
Solution Approach 1:
The patent divides the image reconstruction into two separate pathways: one that applies scatter correction to eliminate noise and artifacts, and another that preserves the original data to maintain resolution. By reconstructing the scatter-only component from scatter-corrected data and combining it with the non-scatter-corrected component, the system achieves noise reduction without the typical resolution loss associated with conventional scatter correction methods.
3Object-generated harmful factors
If conventional scatter correction methods are used, then scatter artifacts are reduced, but noise amplification occurs
Solution Approach 1:
The patent segments the projection data into scatter-affected and non-scatter-affected components, applying scatter correction only to the appropriate portions. The scatter-only component is reconstructed from scatter-corrected projection data, while the non-scatter-corrected component uses the original data. This selective application of scatter correction eliminates the need to amplify noise across the entire image, as the non-scatter-corrected component preserves the original signal-to-noise ratio.
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
This approach improves image quality by reducing noise and artifacts associated with scatter, achieving sufficient noise reduction with minimal resolution degradation, allowing for a compromise between noise reduction and image clarity.
Implementation Method 1
a radiation detector positioned to receive radiation emitted by the radiation source and generate radiation data
Implementation Method 2
The scatter signal detected by the same element also represents the x-rays that are scattered from other x-ray paths into the elements
Data Source
AI summary
A radiological imaging apparatus is provided that includes a radiation source for emitting radiation, a radiation detector positioned to receive radiation emitted by the radiation source and generate radiation data, wherein the radiation data comprises a primary component and a secondary component, and a data processing system. The data processing system is configured to apply image transforms to the primary component using generating functions, build a scatter model basis using the transforms, adjust parameters in the scatter model to fit scatter using the scatter model basis, generate an estimated scatter image by using the fitted scatter model, and modify the radiation data using the scatter image to decrease the scatter in the radiation data thereby generating a scatter corrected image.