Artifact-Reduced X-Ray Image Dataset via Projection Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional algorithms for correcting artifacts in computed tomography images, particularly those caused by small metal objects or asymmetrical objects, are often ineffective and time-consuming, especially in medical procedures involving needles, cannulas, or metallic implants.
Innovation Solution
A method that creates an artifact-reduced x-ray image dataset by identifying object areas in projection datasets, adapting projection values based on the artifact's influence, and generating a new projection dataset to minimize artifact effects, using a processing unit and computer program product for efficient image correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional artifact correction algorithms are used, then artifact reduction is attempted, but processing time increases considerably and effectiveness is limited for small or asymmetrical objects
Solution Approach 1:
The patent segments the correction process by first creating a preliminary artifact-reduced image dataset, then identifying specific object areas in projection datasets, and finally adapting projection values only in those identified regions. This segmented approach avoids applying complex correction algorithms to entire datasets, thereby reducing processing time while maintaining correction effectiveness for small or asymmetrical objects.
Solution Approach 2:
The patent applies local quality by adapting projection values specifically in identified object areas rather than uniformly across the entire dataset. The adaptation is performed as a function of a measure of the influence by the artifact on the projection value, ensuring that correction is applied locally where needed while preserving unaffected data, thus reducing overall processing time.
2Reliability
If complex artifact correction algorithms are applied to entire datasets, then comprehensive correction is achieved, but processing time increases considerably
Solution Approach 1:
The patent extracts and processes only the necessary portions of the data by identifying object areas in projection datasets and adapting projection values only within those specific regions. This extraction approach eliminates the need to process entire datasets with complex algorithms, thereby improving processing efficiency while maintaining correction completeness for affected areas.
Solution Approach 2:
The patent applies partial action by performing artifact correction only on identified object areas rather than on the entire dataset. This partial correction approach is sufficient because it targets only the regions containing artifacts, thereby improving processing efficiency without sacrificing the completeness of artifact reduction in affected areas.
3Measurement precision
If conventional algorithms are used for small metal objects, then some artifact reduction is achieved, but the result is limited and processing remains time-consuming
Solution Approach 1:
The patent performs preliminary action by first creating a preliminary artifact-reduced image dataset before identifying object areas and adapting projection values. This preliminary reduction step improves the quality of subsequent object identification and correction, thereby enhancing measurement precision for small metal objects while keeping the overall processing time reduced through targeted subsequent steps.
Solution Approach 2:
The patent substitutes the conventional mechanical approach of applying uniform correction algorithms with an intelligent system that identifies object areas and adapts projection values based on artifact influence measures. This substitution enables higher measurement precision for small metal objects by using adaptive, data-driven correction rather than blanket algorithmic application.
4Reliability
If artifact correction is applied uniformly across all projections, then comprehensive correction is achieved, but unaffected data is unnecessarily processed
Solution Approach 1:
The patent segments the correction process into preliminary artifact reduction, object area identification, and targeted projection value adaptation. This segmentation ensures that correction coverage is maintained for affected areas while avoiding unnecessary processing of unaffected data, thereby reducing processing complexity without sacrificing correction reliability.
Solution Approach 2:
The patent applies local quality by adapting projection values specifically in identified object areas rather than uniformly across all projections. This localized correction approach maintains artifact correction coverage in affected regions while eliminating unnecessary processing of unaffected data, thereby reducing processing complexity.
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 effectively reduces artifacts in x-ray images, improving image quality and reducing processing time, especially for procedures involving small or asymmetrical objects, by targeting only the affected projections and preserving unaffected data.
Implementation Method 1
a radiation source, for example an x-ray radiation source... recording spatially three-dimensional image data... information about the attenuation of the radiation by the examination object
Data Source
AI summary
A method is for providing a second artifact-reduced x-ray image dataset based on an artifact-affected x-ray image dataset of an examination object, the artifact being caused by an object at least one of on, outside of and within the examination object. In an embodiment, the method includes creating a first artifact-reduced x-ray image dataset based on the artifact-affected x-ray image dataset, based on which a second projection dataset is created; identifying an object area which maps the object in the at least one projection; creating a third projection dataset based on the first projection dataset; and crating the second artifact-reduced x-ray image dataset based on the third projection dataset, through which the second artifact-reduced x-ray image dataset is provided.