Image Reconstruction Using Gradient Reference Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
X-ray image reconstruction in medical imaging often results in uneven exposures due to differences in subject thickness, leading to artifacts like overexposure, which degrade image quality.
Innovation Solution
A system and method for image reconstruction that corrects pixel values in overexposed regions using values from normally exposed regions, employing a processor to segment and correct pixel values based on local gradient references, thereby reducing artifacts and improving image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If direct image reconstruction is performed using acquired imaging data, then the reconstruction process is simple and fast, but image artifacts occur due to uneven exposure and overexposure in certain regions
Solution Approach 1:
The projection image is divided into a first region with normal exposure and a second region with overexposure. This segmentation allows different processing strategies to be applied to different regions, correcting the overexposed areas while preserving the normal areas, thus improving image quality without significantly increasing processing time.
Solution Approach 2:
The patent applies local quality by using pixel values from the normal exposure region to correct pixel values in the overexposed region through gradient reference. This localized correction approach targets only the problematic areas while maintaining the integrity of normal regions, improving overall image quality with minimal additional computational burden.
2Strength
If radiation dose is increased to improve image penetration through thick portions of the subject, then better penetration is achieved, but overexposure artifacts occur in thinner portions of the subject
Solution Approach 1:
The patent converts the harmful overexposure artifacts into beneficial information by using the normal exposure region as a reference to correct the overexposed region. The gradient reference method leverages the relationship between adjacent pixels to reconstruct meaningful data from the overexposed areas, transforming the harmful effect into a useful correction mechanism.
Solution Approach 2:
The patent changes the parameter approach by not uniformly adjusting radiation dose across the entire field of view, but rather processing the acquired data with region-specific correction algorithms. This allows the system to maintain high radiation penetration where needed while computationally correcting overexposed regions, avoiding the need to reduce overall radiation dose.
3Manufacturing precision
If gradient reference method is used to correct pixel values in overexposed regions, then image artifacts are reduced and image quality is improved, but the processing complexity increases
Solution Approach 1:
The patent applies preliminary action by performing gradient calculation and reference pixel selection before the actual correction process. By pre-identifying reference pixels and calculating gradients in the normal exposure region, the system simplifies the subsequent correction steps, reducing the overall processing complexity while maintaining high image quality.
Solution Approach 2:
The patent uses copying by creating a reference model from the normal exposure region that can be applied to correct the overexposed region. The gradient reference method essentially copies the spatial relationships and intensity patterns from the normal region to reconstruct the overexposed area, simplifying the correction process through pattern replication rather than complex iterative optimization.
Data Source
AI summary
The present disclosure relates to a system and a method for image reconstruction. The method may include obtaining a projection image of a subject acquired by an imaging device, the projection image including a first region with a normal exposure corresponding to a first portion of the subject and a second region with an overexposure corresponding to a second portion of the subject; using first pixel values of first pixels in the first region to correct second pixel values of second pixels in the second region; and reconstructing, based on the first pixel values of the first pixels in the first region and the corrected second pixel values of the second pixels in the second region, a target image of the subject.