Virtual Monochromatic X-ray Datasets for Multi-Material Discrimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current X-ray CT systems face challenges in discriminating reference substances greater in number than the types of energies used, due to limitations in data collection and processing techniques.
Innovation Solution
The medical processing apparatus generates virtual monochromatic X-ray datasets by estimating the quantities and mixing ratios of reference materials at multiple positions within a subject, using a processing unit that analyzes X-ray datasets acquired at different energies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dual-energy CT imaging is performed using two types of tube voltages, then reference substance images can be reconstructed based on abundance ratio, but the number of reference substances that can be discriminated is limited to two
Solution Approach 1:
The patent introduces a third dimension by generating virtual monochromatic X-ray datasets at multiple energy levels (e.g., 40, 60, 80, 100, 140, 180 keV) from dual-energy data. This transforms the problem from 2D (two energy levels) to 6D (six energy levels), enabling discrimination of three or more reference substances by adding spectral information dimensions without requiring additional physical X-ray tubes or detectors.
Solution Approach 2:
The patent creates virtual copies of monochromatic X-ray datasets at multiple energy levels through computational processing of dual-energy projection data. These virtual monochromatic datasets serve as synthetic replicas that provide the spectral information needed for multi-material decomposition, avoiding the need for physical multi-energy imaging hardware while achieving enhanced reference substance discrimination capability.
2Measurement precision
If multiple types of X-ray datasets at different energies are acquired, then virtual monochromatic datasets can be generated, but the system complexity increases
Solution Approach 1:
The patent performs preliminary decomposition of dual-energy projection data into basis material components before generating virtual monochromatic datasets. By pre-calculating the abundance ratios of reference substances at each projection angle and then synthesizing monochromatic datasets from these decomposed components, the system simplifies the overall processing pipeline and reduces computational complexity compared to direct multi-energy decomposition.
Solution Approach 2:
The patent uses basis material decomposition as an intermediary step between dual-energy data acquisition and virtual monochromatic dataset generation. The basis material abundances serve as intermediate parameters that mediate the transformation, allowing the system to generate multiple energy-level datasets from two-energy input data through a structured two-stage process that manages computational complexity.
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 discrimination of reference substances exceeding the number of energy types, improving the accuracy and capability of X-ray CT systems in generating various images and diagnosing conditions such as fatty liver.
Implementation Method 1
values each representing absorption of X-rays at each positions in each of the second number of virtual monochromatic X-ray datasets and values each representing absorption of X-rays by each of the second number of reference substances
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A medical processing apparatus (3) includes a processing unit (442,352). Based on a first number of X-ray datasets that correspond to the first number of energies and that are acquired by scanning a subject with X-rays, the processing unit (442,352) generates a second number of virtual monochromatic X-ray datasets, which is a number larger than the first number. Based on the second number of virtual monochromatic X-ray datasets, the processing unit (442,352) estimates quantities of the second number of reference materials mixed or a mixing ratio between the second number of reference materials for each of multiple positions in the subject.