Anisotropic Voxel Image Reconstruction Penalty Interpolation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Anisotropic voxel sizes in PET imaging lead to over-smoothing and step artifacts during image reconstruction, particularly in targeted reconstructions like brain imaging, due to anisotropic penalty computations.

Innovation Solution

The implementation of penalized likelihood (PL) image reconstruction with interpolation of voxel sizes to make them isotropic, allowing for a modified penalty function that reduces anisotropic smoothing and artifacts by interpolating voxel sizes from original to interpolated and back to original for accurate image reconstruction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If anisotropic voxel sizes are used in PET image reconstruction, then the image matrix can represent the physical scanner geometry, but over-smoothing and step artifacts occur during reconstruction

Engineering Contradiction:
Improveimage reconstruction accuracyVSAvoidover-smoothing and step artifacts
Core Design Contradiction:
Manufacturing precisionVSObject-affected harmful factors

Solution Approach 1:

The patent changes the voxel size parameter from anisotropic to isotropic during penalty computation. Specifically, it resamples the image to create isotropic voxels for calculating the penalty function, then applies the computed penalty to the original anisotropic image. This parameter transformation eliminates over-smoothing and step artifacts while preserving the benefits of anisotropic reconstruction.

Inventive Principle:
Principle #35Parameter changes

2Object-affected harmful factors

If isotropic voxel sizes are used for penalty computation, then over-smoothing and artifacts are reduced, but the original anisotropic image structure must be preserved

Engineering Contradiction:
Improveanisotropic smoothing reductionVSAvoidreconstruction process complexity
Core Design Contradiction:
Object-affected harmful factorsVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary isotropic image representation that serves as a bridge between the original anisotropic scanner geometry and the penalty computation. The isotropic resampled image acts as a mediator for calculating the penalty function, while the original anisotropic image structure is preserved and receives the computed penalty. This intermediary approach resolves the contradiction by allowing isotropic penalty computation without permanently altering the anisotropic image structure.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If anisotropic voxels are used in targeted reconstruction, then small lesion contrast recovery is limited, but increasing voxel uniformity improves contrast

Engineering Contradiction:
Improvelesion contrast recoveryVSAvoidvoxel size uniformity
Core Design Contradiction:
Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent temporarily changes the voxel size parameter from anisotropic to isotropic specifically for penalty computation purposes. By resampling to isotropic voxels during penalty calculation, the method improves contrast recovery from small lesions. The original anisotropic voxel dimensions are preserved in the final image, maintaining compatibility with scanner geometry while achieving improved contrast through the isotropic penalty computation.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10482634B2Systems and methods for imaging with anisotropic voxels
Publication Date: 2019.11.19 GE PRECISION HEALTHCARE LLC
  • US10482634B2 patent drawing
  • US10482634B2 patent drawing
  • US10482634B2 patent drawing

AI summary

An imaging system is provided that includes at least one detector configured to acquire imaging information, a processing unit, and a display unit. The processing unit is operably coupled to the at least one detector, and is configured to reconstruct an image using the imaging information. The image is organized into voxels having non-uniform dimensions. The processing unit is configured to perform a penalized likelihood (PL) image reconstruction using the imaging information. The PL image reconstruction includes a penalty function. Performing the penalty function includes interpolating a voxel size in at least one dimension from an original size to an interpolated size before determining a penalty function, determining the penalty function using the interpolated size to provide an initial penalty, interpolating the initial penalty to the original size to provide a modified penalty, and applying the modified penalty in the PL image reconstruction.