Bias Correction Look-Up Table for Low-Count CT Projection Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for bias correction in computed-tomography (CT) image reconstruction, particularly at low-count levels, introduce artifacts due to the logarithmic operation and positivity mapping, which fail to accurately minimize bias and result in suboptimal image quality.
Innovation Solution
A bias-correction look-up table (LUT) is generated to estimate and correct the post-log bias in CT projection data by mapping pre-log data to positive numbers, allowing for pixel-by-pixel bias subtraction and mitigation of artifacts, using statistical models and empirical measurements to account for noise and detection processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If positivity mapping is applied to correct negative intensity values in low-count CT projection data, then the data becomes suitable for logarithm operation and image reconstruction, but bias is introduced resulting in bias artifacts in reconstructed images
Solution Approach 1:
The patent applies preliminary action by performing positivity mapping before the logarithm operation to prevent negative values from causing computational errors. The bias correction LUT is pre-calculated based on statistical models of the positivity mapping process, allowing the bias to be compensated in advance during image reconstruction, thereby preventing bias artifacts in the final image.
Solution Approach 2:
The patent implements feedback by calculating the bias introduced by positivity mapping using statistical models, then using this calculated bias information to correct the projection data through the LUT. This feedback loop continuously refines the correction by comparing the expected bias with the actual bias observed in the data, adjusting the correction factors accordingly to minimize artifacts.
2Object-affected harmful factors
If radiation dose is reduced in CT scans, then patient exposure is minimized, but noise increases to the same order of magnitude as the signal causing negative intensity values
Solution Approach 1:
The patent converts the harmful effect of noise in low-count data into a benefit by developing statistical models that explicitly account for noise characteristics. These models use the noise statistics to calculate appropriate bias correction factors in the LUT, transforming the previously problematic noise into useful information for improving image quality at low radiation doses.
Solution Approach 2:
The patent applies parameter changes by modifying the correction approach based on the count level parameters. The LUT is generated with different correction factors for different count ranges, allowing the system to adapt the positivity mapping and bias correction parameters according to the actual noise conditions in the acquired data, thereby optimizing image quality across varying radiation doses.
3Reliability
If current positivity mapping methods are used to correct negative values, then computational stability is improved, but image quality deteriorates due to inaccurate bias minimization
Solution Approach 1:
The patent applies dynamics by making the bias correction adaptive rather than static. The LUT contains correction factors that are dynamically selected based on the local statistics of the projection data, such as the mean and variance in different regions. This allows the correction method to adapt to varying noise conditions across different parts of the image, improving both computational stability and image quality.
Solution Approach 2:
The patent performs preliminary calculation of bias correction factors based on statistical models of the positivity mapping process. By pre-calculating the expected bias for different input conditions and storing these in the LUT, the system can quickly apply accurate corrections during reconstruction without compromising computational stability or image quality.
Data Source
AI summary
A method and apparatus is provided to obtain projection data representing an intensity of X-ray radiation detected at a plurality of detector elements after traversing an object, the projection data being corrected for a baseline offset, correct the projection data by performing a positivity mapping to generate corrected projection data, perform a logarithm operation on the corrected projection data to generate post-log projection data, correct for a bias of the post-log projection data, using the projection data, to generate bias-corrected projection data, and reconstruct an image of the object from the bias-corrected projection data.