CT Detector Clipping Bias Correction via Mean Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low-dose CT imaging introduces clipping-induced bias artifacts due to electronic noise, causing negative measurements to be clipped and shifting the mean value, which degrades image quality by introducing dark shading artifacts in reconstructed volumetric image data.
Innovation Solution
A system comprising an unlogger, mean estimator, and correction determiner to unlog logged data, estimate the mean value of clipped data, and apply a correction to remove the clipping-induced bias, using a pre-processing circuitry with a clipping-induced bias corrector to produce corrected logged data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If low-dose CT imaging is performed to reduce radiation exposure, then patient safety is improved, but clipping-induced bias artifacts degrade image quality
Solution Approach 1:
The patent converts the harmful clipping effect into a beneficial correction by modeling the clipping-induced bias and subtracting it from the logged data. The clipping artifact, which normally degrades image quality, is now quantified and removed through the correction term β(d,r,v), transforming the harmful effect into a known quantity that can be eliminated.
Solution Approach 2:
The patent changes the parameter space by working with both logged and unlogged data domains. By estimating the mean of unlogged clipped data and transforming it back to the logged domain through the unlogger module, the system operates across multiple parameter spaces to compute the correction, effectively changing the mathematical domain to resolve the contradiction.
2Productivity
If logarithmic operation is applied to convert digitized measurements into attenuation line integrals, then data processing is improved, but clipping of negative values introduces bias
Solution Approach 1:
The patent applies preliminary correction to the logged data before reconstruction. By computing the clipping-induced bias correction term β(d,r,v) and applying it to the logged data in advance, the system prepares corrected input data for the reconstruction algorithm, ensuring that the final image is not degraded by clipping artifacts.
Solution Approach 2:
The patent introduces an intermediary correction process between data acquisition and reconstruction. The clipping-induced bias corrector acts as an intermediary module that receives logged data, computes the correction based on unlogged clipped data statistics, and outputs corrected logged data, mediating between the logarithmic transformation and the final image reconstruction.
3Ease of operation
If negative measurements are clipped to small positive values, then logarithmic operation becomes defined, but mean value shifts causing dark shading artifacts
Solution Approach 1:
The patent implements feedback by using the estimated mean of unlogged clipped data to compute the clipping-induced bias correction. The system measures the effect of clipping through the mean estimation, feeds this information back through the unlogger and correction determiner, and applies the resulting correction to eliminate the mean shift, creating a closed-loop correction mechanism.
Data Source
AI summary
A system (116) includes an unlogger (202) configured to unlog logged data, to produce unlogged clipped data. The logged data includes attenuation line integrals and clipping-induced bias. The system further includes a mean estimator (204) configured to estimate a mean value of the unlogged clipped data. The system further includes a correction determiner (206) configured to determine correction to the clipping-induced bias based on the estimated mean value of the unlogged clipped data. The system further includes an adder (210) configured to correct the logged data with the correction to produce corrected logged data.