Tomographic Reconstruction Algorithm for X-Ray Cone-Beam Scan Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current algorithms for reconstructing cone-beam x-ray scan data in tomographic imaging, such as the FDK algorithm, suffer from severe artifacts in short scan scenarios, particularly in C-arm CT applications, and are not optimal for data without extrapolation, especially outside the central plane.

Innovation Solution

A novel tomographic reconstruction algorithm using shift-invariant filtering and backprojection with a 1D Hilbert transform is proposed, which combines conventional FDK reconstruction with differential backprojection to suppress artifacts, allowing for optimal reconstruction on a short scan comparable to a full scan, even with reduced data acquisition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If FDK algorithm is used for short scan reconstruction, then computation efficiency is improved through 1D shift-invariant filtering, but severe cone-beam artifacts are introduced

Engineering Contradiction:
Improvecomputation efficiencyVSAvoidcone-beam artifacts
Core Design Contradiction:
ProductivityVSObject-generated harmful factors

Solution Approach 1:

The patent applies the Hilbert transform to convert the harmful cone-beam artifacts into beneficial correction terms. The Hilbert transform of the FDK reconstruction produces a correction image that, when subtracted from the original FDK reconstruction, eliminates the artifacts while preserving the computational efficiency of the 1D shift-invariant filtering structure.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent introduces an intermediary correction term based on the Hilbert transform between the FDK reconstruction and the final result. This intermediary correction image acts as a mediator that removes the harmful artifacts without requiring a complete redesign of the reconstruction algorithm, thus maintaining computational efficiency while eliminating cone-beam artifacts.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Quantity of substance

If Parker's weighting modification is applied to FDK, then data redundancy is handled, but reconstruction accuracy is limited to mid-plane only

Engineering Contradiction:
Improvedata redundancy handlingVSAvoidreconstruction accuracy
Core Design Contradiction:
Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent extends the correction approach from the mid-plane (2D) to the entire 3D volume by applying the Hilbert transform in the third dimension (z-direction). This dimensional extension allows the correction to be applied uniformly across all slices, achieving accurate reconstruction throughout the entire volume rather than just the mid-plane.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Loss of time

If short scan data acquisition is used, then examination time is reduced, but reconstruction quality deteriorates compared to full scan

Engineering Contradiction:
Improveexamination timeVSAvoidreconstruction quality
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent performs preliminary action by pre-calculating the Hilbert transform correction term from the FDK reconstruction. This preliminary correction calculation allows the short scan reconstruction to achieve full-scan quality by applying the correction after the initial FDK reconstruction, thus compensating for the reduced data acquisition without requiring additional scan time.

Inventive Principle:
Principle #10Preliminary action

4Manufacturing precision

If mathematical exact algorithm is applied to short scan, then data redundancy is handled exactly, but reconstruction optimality is not achieved

Engineering Contradiction:
Improvedata redundancy handling accuracyVSAvoidreconstruction optimality
Core Design Contradiction:
Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The patent changes the parameter of the reconstruction algorithm by introducing the Hilbert transform operation. This parameter change transforms the reconstruction from a mathematically exact but non-optimal solution to an optimal solution that achieves both exact data redundancy handling and optimal reconstruction quality, as demonstrated by the elimination of cone-beam artifacts and improvement in image quality metrics.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS7409033B2Tomographic reconstruction for x-ray cone-beam scan data
Publication Date: 2008.08.05 THE BOARD OF TRUSTEES OF THE LELAND STANFORD JUNIOR UNIV
  • US7409033B2 patent drawing
  • US7409033B2 patent drawing
  • US7409033B2 patent drawing

AI summary

Disclosed is x-ray cone beam scan data reconstruction of an imaged object with a reconstruction algorithm using shift invariant filtering and backprojection with the maximum tomographic capability of a circular scan larger than p plus cone angle, when CB data is not truncated and data extrapolation is not allowed. The reconstruction scheme includes a conventional FDK reconstruction and a parallel reconstruction using differential back projection and 1D Hilbert transform to suppress the CB artifacts.