Computed Tomography Volume Data Fusion for Artefact Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computed tomography methods struggle to consistently produce high-quality volume data and identify critical regions due to variations in reconstruction techniques and conditions, leading to potential artefacts and unreliable data quality.
Innovation Solution
Reconstructing two sets of volume data under different conditions, calculating individual confidence measures for each voxel, and using these measures to identify and visualize critical regions through conflict analysis and data fusion, allowing for improved data quality and artefact suppression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If different reconstruction techniques are used, then various advantages can be obtained under different conditions, but different reconstructed density values are yielded in certain volume regions
Solution Approach 1:
The patent combines multiple reconstruction techniques (analytical methods like Feldkamp, algebraic methods like ART/SART, and statistical methods like maximum likelihood reconstruction) into a unified framework that processes projection data through multiple techniques and integrates their results. This merging approach allows the system to leverage the strengths of each technique while maintaining consistent reconstructed density values across different volume regions.
Solution Approach 2:
The patent creates a universal reconstruction system that can adaptively apply different reconstruction techniques based on the specific acquisition conditions and sample characteristics. The system evaluates multiple techniques and selects or combines them appropriately, making the reconstruction process universally applicable to various CT scanning scenarios while maintaining density value consistency.
2Device complexity
If a single reconstruction method is used, then the process is simple, but it cannot provide comprehensive advantages under all conditions and for all sample types
Solution Approach 1:
The patent implements a dynamic reconstruction approach where the system automatically adjusts the reconstruction method or combination of methods based on the specific acquisition conditions, sample type, and detected artefacts. This dynamic adaptation allows the system to maintain relative simplicity in operation while achieving high versatility across different reconstruction scenarios.
Solution Approach 2:
The system incorporates feedback mechanisms that evaluate the quality of reconstructed images and detected artefacts, then use this information to adjust the reconstruction parameters or select appropriate techniques. This feedback loop enables the system to automatically optimize the reconstruction process for different conditions without requiring complex manual intervention.
3Ease of operation
If traditional reconstruction methods are used, then the reconstruction process is straightforward, but critical regions with questionable data quality cannot be identified
Solution Approach 1:
The patent introduces confidence measures as an intermediary assessment layer that evaluates the quality of reconstructed voxels without complicating the basic reconstruction process. These confidence measures act as a mediator between the reconstruction process and quality assessment, providing reliable data quality information while maintaining the simplicity of the underlying reconstruction operations.
Solution Approach 2:
The patent replaces manual quality assessment with an automated confidence measure calculation system that objectively evaluates data quality. This substitution eliminates the need for operator intervention in quality assessment while providing comprehensive and reliable identification of critical regions with questionable data quality.
Data Source
AI summary
A method of using computed tomography for determining a volumetric representation of a sample, involving a first reconstruction for reconstructing first reconstructed volume data of the sample from first x-ray projection data of the sample taken by an x-ray system, a second reconstruction for reconstructing second reconstructed volume data of the sample from second x-ray projection data of the sample taken by an x-ray system, characterized by calculating first individual confidence measures for single voxels of the first reconstructed volume data, calculating second individual confidence measures for single voxels of the second reconstructed volume data, and calculating, in a subsequent step, at least one resulting set of individual values for each voxel based on the first individual confidence measures and the second individual confidence measures.