Multi-Neighbor Correction for Source Leakage in Dual-Energy CT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-ideal source switching in dual-energy CT scanners leads to source leakage, causing data contamination and image artifacts due to overlap between different energy spectrum radiation, which diminishes image quality.
Innovation Solution
A multi-neighbor correction method is applied to projection space data to correct for source leakage by using correction factors determined from phantom projection space data, subtracting contaminated data from adjacent views to isolate pure energy spectrum data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If source switching is used to obtain multiple energy spectra, then measurement precision for atomic number is improved, but source leakage causes data contamination and image artifacts that worsen image quality
Solution Approach 1:
The patent applies preliminary action by using phantom projection data acquired during the scan to calculate correction factors before correcting the actual object projection data. The correction factors are computed in advance based on the measured source leakage characteristics from the phantom, then applied to remove contamination from the object data, thereby resolving the contradiction between obtaining multiple energy spectra and eliminating source leakage artifacts
Solution Approach 2:
The patent converts the harmful source leakage into a beneficial correction opportunity. By intentionally measuring the source leakage through phantom scans and using it to generate correction factors, the harmful contamination is transformed into useful information that enables precise correction of the object data, thereby improving both measurement precision and image quality
2Productivity
If voltage switching speed is increased to reduce overlap, then productivity is improved, but non-ideal switching causes data contamination that worsens measurement precision
Solution Approach 1:
The patent implements feedback by using the actual measured source leakage from phantom projection data to compute correction factors that are then applied to the object data. This closed-loop approach continuously adapts to the actual switching conditions and leakage characteristics, enabling high-speed scanning while maintaining energy spectrum purity through data-driven correction
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
The multi-neighbor correction significantly reduces artifacts and improves image quality by effectively isolating and correcting for source leakage, enhancing the accuracy of atomic number measurements and image clarity.
Implementation Method 1
While the sources are switching the energy spectrum of the radiation, overlap between x-ray emissions may occur. Such overlap is hereafter referred to as source leakage
Data Source
AI summary
The techniques described herein provide for correcting projection data that comprises contamination due to source switching in a multi energy scanner. The correction is a multi-neighbor correction. That is, it uses data from at least two other views of an object (e.g., generally a previous view and a subsequent view) to correct a current view of the object. The multi-neighbor correction may use one or more correction factors to determine how much data from the other two views to use to correct the current view. The correction factor(s) are determined based upon a calibration that utilizes image space data and/or projection space data of a phantom. In this way, the correction factor(s) account for source leakage that occurs in multi energy scanners.