Tilted Helical CT Image Reconstruction Without Data Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional helical reconstruction methods for tilted helical scan in CT devices introduce artifacts due to data correction, which degrades the quality of reconstructed images.
Innovation Solution
An image reconstruction method and device that calculates the ray channel parameter and slice number for each pixel position using geometric relationships, allowing for image reconstruction via a three-dimensional helical back-projection method without data correction, thereby avoiding artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional helical reconstruction method is used for tilted helical scan, then image reconstruction can be performed, but artifacts appear and image quality deteriorates
Solution Approach 1:
The patent changes the reconstruction parameters by calculating ray channel parameters and slice numbers that account for the tilted scan geometry. Instead of using conventional reconstruction parameters designed for non-tilted scans, the method adapts the parameters to match the actual tilted helical trajectory, thereby eliminating artifacts and improving image quality.
Solution Approach 2:
The patent transitions from conventional two-dimensional slice reconstruction to three-dimensional helical back-projection reconstruction. By incorporating the third dimension (helical trajectory) and calculating spatial coordinates that reflect the tilted scan geometry, the method eliminates artifacts caused by mismatched reconstruction assumptions.
2Object-generated harmful factors
If data correction such as interpolation or smoothing is performed before image reconstruction, then artifacts may be reduced, but raw data is contaminated and reconstruction precision deteriorates
Solution Approach 1:
The patent performs preliminary calculation of ray channel parameters and slice numbers based on the known tilted scan geometry before reconstruction. By pre-computing the correct spatial mapping parameters, the method eliminates the need for subsequent data correction steps like interpolation or smoothing, thus preserving raw data accuracy while preventing artifacts.
Solution Approach 2:
The patent replaces the mechanical data correction process (interpolation, smoothing) with a mathematical coordinate transformation approach. Instead of physically adjusting the data to fit conventional reconstruction assumptions, the method transforms the coordinate system to match the actual tilted scan geometry, eliminating artifacts without contaminating the raw data.
3Adaptability or versatility
If tilted helical scan mode is used to reduce radiation damage and improve curvilinear structure imaging, then scanning effectiveness is improved, but conventional reconstruction methods produce artifacts
Solution Approach 1:
The patent introduces dynamic parameter calculation that adapts to the tilted scan geometry. The ray channel parameters and slice numbers are calculated based on the actual dynamic trajectory of the tilted helical scan, allowing the reconstruction method to accommodate the adapted scanning mode while maintaining high image quality without artifacts.
Data Source
AI summary
Image reconstruction method and device for tilted helical scan are provided. The method includes: acquiring data generated by the tilted helical scan; calculating a ray channel parameter and a slice number corresponding to each pixel position based on a geometric relationship that a ray of the tilted helical scan emitting from a ray source to a detector intersects with a pixel; and performing image reconstruction by using a three-dimensional helical back-projection method based on the acquired data, and the ray channel parameter and the slice number corresponding to each pixel position. With the method and device, there is no need to correct the raw data, which ensures the accuracy of the data, and the artifact is avoided, which ensures the accuracy of the reconstructed image.