CT Image Correction via Motion Artifact Parameter Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current CT imaging technologies face challenges in accurately correcting motion artifacts caused by patient movement during computed tomography (CT) scans, leading to blurred images and reduced diagnostic efficiency.
Innovation Solution
A method and apparatus for measuring parameters such as sharpness and entropy values in CT images, calculating a correction possibility, and determining whether to perform correction based on these values, which includes recognizing motion artifact positions and intensities, and using machine learning to update correction equations for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If motion artifacts are corrected using related art techniques (identifying ROI, measuring sensor signals, estimating patient motion), then correction can be performed, but the accuracy of correction is low and unnecessary operations degrade efficiency
Solution Approach 1:
The patent extracts and measures specific parameter values (sharpness, entropy, edge position) directly from the X-ray projection images and reconstruction images to quantify motion artifacts. This extraction approach eliminates the need for complex sensor-based motion estimation and ROI identification, achieving both high correction accuracy and efficiency by focusing only on the essential image quality parameters
Solution Approach 2:
The system uses the image data itself to determine correction necessity and parameters. By measuring sharpness values, entropy values, and edge position changes within the images, the system self-evaluates the presence and severity of motion artifacts without requiring external sensors or manual ROI specification, thereby improving both accuracy and operational efficiency
2Reliability
If correction is performed on all CT images, then motion artifacts may be removed, but unnecessary operations are performed on images that do not require correction
Solution Approach 1:
The patent applies partial action by selectively performing correction only on images that meet specific criteria. The system measures parameter values and compares them against reference values or thresholds, applying correction only when the measured values indicate the presence of motion artifacts. This avoids unnecessary processing of images that already have acceptable quality
Solution Approach 2:
The system implements feedback by measuring parameter values from the images, comparing these measurements against reference values, and using the comparison results to determine whether correction should be applied. This feedback mechanism ensures that correction is performed only when and where it is actually needed, optimizing both image quality and processing efficiency
Data Source
AI summary
A method and apparatus correct a computed tomography (CT) image with motion artifacts. The method of correcting a CT image may include: obtaining a reconstruction image of an object by reconstructing an X-ray projection image; measuring a parameter value related to motion artifacts that occur due to movement of the object in at least one of the X-ray projection image or the reconstruction image; calculating a correction possibility for the reconstruction image based on the measured parameter value; and determining whether to perform correction on the reconstruction image based on the calculated correction possibility.