X-ray CT Image Reconstruction Using Weighted Iterative Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing X-ray CT systems face challenges in rapidly converging images when anomalous data is detected from an X-ray detecting element, leading to artifacts and prolonged processing times due to the exclusion of faulty data in iterative processing.
Innovation Solution
An X-ray CT apparatus that stores scan data from multiple detecting elements, identifies anomalous data, and applies weighted processing using a projection matrix to rapidly converge pixel data, allowing for the inclusion of adjacent data to correct image artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If anomalous detection data from fault X-ray detecting elements is excluded from iterative processing, then image artifacts are prevented, but convergence of iterative processing becomes slower
Solution Approach 1:
The patent applies local quality by assigning different weights to different detection data based on their reliability. Specifically, detection data from detecting elements adjacent to fault elements are given higher weights, while data from fault elements themselves are given lower or zero weights. This localized weighting approach allows the system to utilize as much data as possible for faster convergence while preventing artifacts from faulty elements, thus resolving the contradiction between image quality and processing speed.
Solution Approach 2:
The patent changes the parameter of data weighting in the iterative processing system. By dynamically adjusting weights assigned to detection data based on the spatial relationship with fault elements, the system optimizes the contribution of each data point. This parameter change enables rapid convergence by emphasizing reliable data while suppressing problematic data, achieving both fast processing and high image quality.
2Reliability
If iterative processing is performed without using detection data from fault X-ray detecting elements, then artifacts are avoided, but the number of iterations required increases
Solution Approach 1:
The patent implements local quality by applying spatially-dependent weighting to detection data. Detecting elements adjacent to fault elements receive higher weights, allowing their data to be utilized more aggressively in the iterative process. This localized approach to data quality assessment enables the system to maintain image accuracy while reducing the number of iterations needed for convergence, thereby improving processing efficiency.
Solution Approach 2:
The patent applies partial action by selectively including or excluding detection data based on their proximity to fault elements. Rather than completely excluding all data from regions near faults or using all data uniformly, the system partially includes data with appropriate weighting. This selective approach allows sufficient data to be used for efficient processing while preventing excessive influence from faulty data, thus resolving the contradiction between accuracy and efficiency.
3Productivity
If detection data from all X-ray detecting elements is used uniformly in iterative processing, then processing speed increases, but image artifacts occur due to anomalous data
Solution Approach 1:
The patent applies local quality by assigning spatially-varying weights to detection data based on their relationship with fault elements. Elements adjacent to faults receive higher weights, fault elements receive lower or zero weights, and this creates a localized quality map that guides the iterative processing. This approach enables the system to process data efficiently while maintaining image quality by preventing artifact propagation from faulty elements.
Solution Approach 2:
The patent introduces dynamics by making the weight assignment adaptive rather than static. The weighting scheme dynamically adjusts based on the identified fault element positions, allowing the system to optimize processing for each specific fault configuration. This dynamic approach enables fast processing speed while adaptively preventing artifacts, resolving the contradiction between productivity and reliability.
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 enables the generation of artifact-free X-ray CT images by accelerating the convergence of iterative operations, improving image reliability and reducing processing time compared to previous methods.
Implementation Method 1
an X-ray source that x-irradiates a subject and an X-ray detector that is arranged at a position facing to the X-ray source to detect X-rays passing through the subject
Implementation Method 2
an X-ray detector that is arranged at a position facing to the X-ray source to detect X-rays passing through the subject
Data Source
AI summary
An X-ray detector (320), configured to have X-ray detecting elements (322) arranged in array, detects the intensity of X-rays that are radiated from the X-ray tube (311) and have passed through a subject (500). A data processing device (420) executes steps of: arranging scan data, which is based on the intensity of X-rays, in an array sequence of the X-ray detecting elements (322) or in a time sequence, to detect an anomalous scan data; associating a weight greater than “1” with scan data adjacent to the anomalous scan data, and associating a weight of “1” with other imaging data; calculating an update amount of a pixel vector that reflects these weights; and performing an iterative operation using the update amount to generate an X-ray CT image of the subject (500).