Portable X-Ray CT Shot Geometry Self-Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional geometric calibration techniques for portable x-ray CT are not suitable for open-configuration systems, requiring a reference constellation to remain in the scene, which is not mechanically repeatable, making it difficult to determine shot geometry accurately.
Innovation Solution
A method for self-determining shot geometry in open-configuration portable x-ray CT using an unknown constellation, involving the estimation of an unknown constellation model and reconstructing volume images using computed tomography, with a customized geometric calculation technique for a flat panel digital detector and cone beam x-ray source.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional geometric calibration techniques are used, then shot geometry can be determined accurately, but a reference constellation must remain in the scene which is not suitable for portable open-configuration systems
Solution Approach 1:
The system performs self-calibration by automatically determining shot geometry from the captured image data itself, without requiring external reference constellations or manual calibration procedures. The geometry estimation algorithm extracts geometric information directly from the x-ray shot images, enabling the portable system to calibrate itself in the field.
Solution Approach 2:
The patent replaces mechanical calibration systems (reference constellations, rotation stages) with a computational approach. Instead of using physical reference objects and mechanical repeatability, the system uses algorithms to estimate shot geometry from the imaging data, substituting mechanical determination with mathematical computation.
2Measurement precision
If a reference constellation is used for geometric calibration, then shot geometry can be determined, but the constellation must be identified and tracked in every x-ray shot which is difficult to perform
Solution Approach 1:
The patent extracts geometric information directly from the x-ray shot images without requiring separate identification and tracking of fiducial markers. The geometry estimation algorithm processes the image data to determine source and detector positions and orientations, eliminating the need for fiducial identification and tracking steps.
Solution Approach 2:
The system determines shot geometry selflessly from the image data itself, without requiring additional processing steps for fiducial identification. The algorithm automatically extracts geometric parameters from the captured x-ray shots, making the process simpler and more automated.
3Measurement precision
If mechanical repeatability is assumed, then shot geometry can be determined using rotation stage calibration, but portable systems do not have mechanical repeatability
Solution Approach 1:
The patent replaces mechanical calibration approaches with a computational geometry estimation method. Instead of relying on mechanical repeatability of rotation stages, the system uses algorithms to determine shot geometry from the imaging data, making it applicable to portable systems without precise mechanical repeatability.
Solution Approach 2:
The system changes from fixed mechanical parameters (rotation stage calibration) to flexible computational parameters (geometry estimation from images). This allows the system to adapt to portable configurations where mechanical repeatability cannot be guaranteed, by determining geometry parameters directly from each shot rather than assuming repeatability.
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
Enables accurate determination of x-ray shot geometry without mechanical repeatability, allowing for volume image reconstruction and reducing image artifacts, even when not all fiducials are visible in every shot.
Implementation Method 1
cone beam x-ray source
Implementation Method 2
reconstructing a volume image from the one or more captured shot images and the one or more geometries using computed tomography
Data Source
AI summary
Self-determined shot geometry for open-configuration portable x-ray computed tomography (CT) includes capturing one or more x-ray shot images into a computing system, and estimating an unknown constellation model and one or more shot geometries. It also includes reconstructing a volume image from the one or more captured shot images and the one or more geometries using computed tomography.