Helical CT Streak Artifact Reduction via Adaptive Upsampling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Helical Cone Beam Computer Tomography (CT) systems suffer from streak artifacts due to the singularity of cone beam weighting functions at Z-positions near the image plane, especially when the rotation is two or more, leading to inaccuracies and noise in image reconstruction.

Innovation Solution

The implementation of a multi-slice X-ray CT apparatus that uses upsampling to add data values for hypothetical rows, thereby stabilizing the cone beam weighting function and reducing streak artifacts by interpolating additional data points near the image plane, particularly using a predetermined number and width of additional data points determined by parameters like Ufactor and Uview.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If cone beam weighting function is used with rotation equal to or more than two, then detector utilization is improved, but streak artifacts are generated due to singularity at Z-positions near the image plane

Engineering Contradiction:

Improvedetector utilizationVSAvoidstreak artifacts

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent applies preliminary action by performing upsampling of projection data before the cone beam weighting step. By interpolating additional data points in advance, the method ensures that sufficient data is available to compute accurate weighting values, preventing the singularity problem that causes streak artifacts while maintaining high detector utilization

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If image plane projects to a small fraction of the detector row, then reconstruction accuracy is improved, but CBW weight function becomes narrower than detector size causing inaccuracies

Engineering Contradiction:

Improvereconstruction accuracyVSAvoidweight function accuracy

Core Design Contradiction:

Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent applies dimensionality change by upsampling the projection data in the detector row dimension. When the image plane projects to a small fraction of the detector row, the method interpolates additional data points along the detector row direction, effectively increasing the sampling density in that dimension to maintain accurate weight function computation

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Object-affected harmful factors

If non-linear form of uCBW is used, then cone beam artifacts are reduced, but inaccuracies cannot be accurately determined for all image pixels

Engineering Contradiction:

Improvecone beam artifactsVSAvoidweight determination accuracy

Core Design Contradiction:

Object-affected harmful factorsVSMeasurement precision

Solution Approach 1:

The patent applies preliminary action by performing upsampling before the non-linear uCBW computation. By pre-interpolating the projection data to increase sampling density, the method ensures that the non-linear weight function can be accurately evaluated for all image pixels, including those that would otherwise fall between original data points

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS8724876B2Method and system for substantially reducing streak artifacts in helical cone beam computer tomography (CT)

Publication Date: 2014.05.13 TOSHIBA MEDICAL SYST CORP

AI summary

Streak artifacts arise in helical CT reconstruction with cone beam weighting (CBW) with helical pitch ratio between 0.5 and 1.0 in a prevalent 2PI mode. The sreak artifacts are substantially removed by applying upsampling to the measured data in the segment direction before weighting. Furthermore, by making the upsampling adaptive to the view Z-position, an amount of extra processing is greatly reduced to near 1%.