Iterative Energy Bin Optimization for Spectral CT Image Quality

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing photon-counting spectral computed tomography technologies lack an effective method to optimize energy bin parameters adaptively for specific computed tomography applications, leading to suboptimal image quality and segmentation of anatomical structures.

Innovation Solution

A method involving iterative steps that adjust energy bin parameters using a machine learning-based approach, incorporating reconstruction, segmentation, and evaluation algorithms to optimize energy bin settings based on photon-counting spectral computed tomography data, allowing for adaptive optimization of energy bin parameters for improved image quality and segmentation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If the same pre-defined window parameters of the energy bins are used for different kinds of computed tomography applications, then the device complexity is reduced, but the image quality and segmentation quality deteriorate

Engineering Contradiction:
Improveenergy bin parameter configurationVSAvoidimage quality and segmentation quality
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic adaptation of energy bin parameters through iterative optimization. The system starts with pre-defined parameters and progressively adjusts them based on evaluation metrics (image quality and segmentation quality) until optimal parameters are found. This dynamic approach resolves the contradiction by allowing parameters to change from static/pre-defined to adaptive/optimized.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent employs feedback mechanisms where evaluation algorithms assess the quality of images and segmentations produced by segmentation algorithms, and this evaluation data feeds back into optimizing the energy bin parameters. This closed-loop feedback system enables continuous improvement of parameters based on actual performance, resolving the contradiction between simplicity and quality.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If iterative optimization steps are performed to adapt energy bin parameters to specific computed tomography applications, then the image quality and segmentation quality are improved, but the processing time and computational complexity increase

Engineering Contradiction:
Improveimage quality and segmentation qualityVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by using pre-defined window parameters as initial values before the iterative optimization process. This preliminary setup provides a starting point that is already reasonable, reducing the number of iterations needed to reach optimal parameters and thus reducing processing time while maintaining quality improvement.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent applies partial action by performing a limited number of iterative optimization steps rather than exhaustive optimization. The system performs enough iterations to achieve significant quality improvement but stops before excessive computational effort is expended, balancing quality enhancement with time constraints.

Inventive Principle:
Principle #16Partial or excessive action

3Adaptability or versatility

If adaptive optimization of energy bin parameters is implemented, then the utilization of spectral information is improved, but the device complexity and algorithm complexity increase

Engineering Contradiction:
Improvespectral information utilizationVSAvoidoptimization system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements multi-functionality by designing an optimization system that can handle multiple computed tomography applications and multiple types of evaluations (image quality assessment, segmentation quality assessment). This universal optimization framework resolves the contradiction by providing a single adaptable system that improves spectral information utilization across different applications without proportionally increasing complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3550517B1Method and providing unit for providing an optimized energy bin parameter set for photon-counting spectral computed tomography
Publication Date: 2020.08.05 SIEMENS HEALTHCARE GMBH
  • EP3550517B1 patent drawingFigure 1~2
  • EP3550517B1 patent drawingFigure 3~4

AI summary

In one aspect the invention relates to a method for providing an optimized energy bin parameter set for photon-counting spectral computed tomography, the method comprising the following steps: - receiving photon-counting spectral computed tomography data related to a plurality of energy bins and an initial energy bin parameter set, - performing iteration steps of a plurality of iteration steps, - wherein the input of the first iteration step of the plurality of iteration steps comprises the initial energy bin parameter set as an input energy bin parameter set, - wherein the input of each further iteration step of the plurality of iteration steps comprises an adjusted energy bin parameter set calculated in the preceding iteration step of the plurality of iteration steps as the input energy bin parameter set.