Intelligent Trajectory Optimization for X-ray CT Reconstruction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In X-ray CT systems, transparent test objects pose challenges in determining precise projection geometry due to overlaid object structures, leading to potential errors in three-dimensional reconstruction, especially when using markers that may not be recognizable.

Innovation Solution

A method is developed to optimize the trajectory of a test part within the X-ray CT system by intelligently varying it based on object orientation, using a parameterizable trajectory and an optimization algorithm to minimize a quality function that accounts for material translucency, allowing for improved reconstruction quality even with low-precision manipulators and six-axis industrial robots.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If markers are used to determine projection geometry, then the position can be determined, but markers may be overlaid by object structures and become unrecognizable leading to errors

Engineering Contradiction:
Improveprojection geometry determinationVSAvoidmarker recognition
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces an intelligent trajectory optimization system as an intermediary between the manipulator and the reconstruction process. Instead of relying on markers that may be obscured, the system uses software-based trajectory optimization that adapts to the actual object geometry, thereby mediating the determination of projection geometry without requiring visible markers.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/marker-based system for determining projection geometry with a computational approach. The intelligent trajectory optimization algorithm substitutes the physical marker system, using mathematical optimization and software processing to determine the necessary projection geometries without relying on physical markers that could be obscured.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Ease of manufacture

If a fixed trajectory is used for manipulation, then the process is simple, but the reconstruction quality may be insufficient for transparent objects

Engineering Contradiction:
Improvetrajectory implementationVSAvoidreconstruction quality
Core Design Contradiction:
Ease of manufactureVSManufacturing precision

Solution Approach 1:

The patent transforms the static, fixed trajectory into a dynamic, adaptive trajectory. The intelligent optimization algorithm continuously adjusts the manipulation trajectory based on the specific object characteristics and transparency properties, making the trajectory flexible and adaptive rather than rigid and predetermined.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameters of the trajectory through intelligent optimization. The system varies trajectory parameters such as projection angles, positions, and spacing to optimize the quality function specifically for transparent objects, thereby improving reconstruction quality without significantly complicating the implementation.

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If the trajectory is optimized iteratively, then the reconstruction quality improves, but the optimization time increases

Engineering Contradiction:
Improvereconstruction qualityVSAvoidoptimization time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by pre-defining a reasonable trajectory parameter space and using efficient optimization algorithms that converge quickly. The system prepares the optimization framework in advance with appropriate constraints and initial conditions, reducing the computational burden during actual optimization and minimizing time loss while maintaining high reconstruction quality.

Inventive Principle:
Principle #10Preliminary action

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 enhances the quality of reconstructed volumes by optimizing the acquisition trajectory, ensuring accurate determination of projection geometry and improving image quality, even with less precise manipulator positioning, and reduces the time required for optimization by limiting iteration steps.

Implementation Method 1

an X-ray source with focus, an X-ray detector... has, in order to generate recordings of the test part in different positions

Methodology Applied
Scientific EffectX-ray transmission: X-Ray

Implementation Method 2

a manipulator that moves the test part within the X-ray CT system, has, in order to generate recordings of the test part in different positions, the manipulator follows a predeterminable trajectory

Methodology Applied
Scientific EffectMechanical motion:

Data Source

PatentEP3274962B1Method for the reconstruction of a test part in an x-ray ct method in an x-ray ct system by means of an intelligent path curve
Publication Date: 2019.03.06 COMET YXLON GMBH
  • EP3274962B1 patent drawingFigure 1~2
  • EP3274962B1 patent drawingFigure 3~4
  • EP3274962B1 patent drawingFigure 5~6

AI summary

The invention relates to a method for the reconstruction of a test part in an x-ray CT method in an x-ray CT system, which has an x-ray source having focus, an x-ray detector, and a manipulator, which moves the test part within the x-ray CT apparatus, wherein, to generate recordings of the test part in various positions of the manipulator, a predefinable configurable path curve is traveled and makes the recordings at triggered positions, wherein the position of the manipulator for each recording is determined and, from this, the respective associated projection geometry is calculated. Then, a value of a quality function is calculated for this path curve and, after that, a further path curve is followed which has parameters different from the preceding path curve, while the generation of further recordings of the test part is carried out at the triggered positions and the value of the quality function is calculated for this purpose, the last-named step is repeated, the path curve is determined iteratively by means of an optimization algorithm at the value of which the quality function is minimal, further recordings of further test parts that are same as the said test part are created along the path curve at which the minimum of the quality function has been established previously, and for each test part, by using the assignment of the individual recordings to the respective projection geometry, a CT reconstruction is carried out by means of a suitable algorithm.