Homogeneous Deformation Approximation for CT Bone Subtraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Non-rigid registration in CT bone subtraction for lytic bone lesions can result in incorrect shrinkage or expansion of bone marrow edema patterns, reducing contrast between affected and normal bone marrow regions, while rigid registration may enhance contrast but increase subtraction artifacts.

Innovation Solution

A medical image processing system that performs non-rigid registration to determine a deformation field, identifies approximately homogeneous regions, and applies a substantially homogeneous approximation to the deformation field within these regions, replacing non-rigid transformations with rigid or affine approximations to improve image alignment and subtraction accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If non-rigid registration is used to align image data sets, then image alignment accuracy is improved, but contrast between bone marrow edema and normal regions is reduced

Engineering Contradiction:

Improveimage alignment accuracyVSAvoidcontrast between bone marrow edema and normal regions

Core Design Contradiction:

Measurement precisionVSIllumination intensity

Solution Approach 1:

The patent segments the image data into different anatomical regions (bone, soft tissue, etc.) and applies different registration strategies to each region. Rigid registration is applied to bone regions to preserve contrast, while non-rigid registration is applied to soft tissue regions to accommodate deformation, thus resolving the contradiction between alignment accuracy and contrast preservation

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies different registration transformations to different spatial locations within the image. Specifically, rigid transformations are applied to bone regions where contrast must be preserved, while non-rigid transformations are applied to soft tissue regions where anatomical deformation occurs, achieving both accurate alignment and contrast preservation in their respective domains

Inventive Principle:

Principle #3Local quality

2Illumination intensity

If rigid registration is used to align image data sets, then contrast between bone marrow edema and normal regions is improved, but subtraction artifacts increase

Engineering Contradiction:

Improvecontrast between bone marrow edema and normal regionsVSAvoidsubtraction artifacts

Core Design Contradiction:

Illumination intensityVSObject-generated harmful factors

Solution Approach 1:

The patent divides the image into bone and soft tissue segments, applying rigid registration only to bone regions to maintain contrast while limiting artifact generation to non-critical areas, whereas soft tissue regions use non-rigid registration to reduce artifacts from anatomical deformation

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements spatially varying registration quality by applying high-contrast-preserving rigid transformations to bone regions and artifact-reducing non-rigid transformations to soft tissue regions, optimizing the trade-off between contrast and artifact reduction in different anatomical locations

Inventive Principle:

Principle #3Local quality

3Measurement precision

If non-rigid registration is used to correct motion, then image alignment is improved, but bone marrow edema pattern accuracy is reduced

Engineering Contradiction:

Improveimage alignmentVSAvoidbone marrow edema pattern accuracy

Core Design Contradiction:

Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent segments the image data into bone and soft tissue regions, applying non-rigid registration only to soft tissue to correct motion artifacts while preserving bone structure integrity, thus maintaining BMEP accuracy in bone regions while still achieving motion correction in movable soft tissue regions

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies different transformation qualities to different anatomical structures: rigid transformations preserve the precise geometry of bone and BMEP patterns, while non-rigid transformations accommodate soft tissue motion, achieving motion correction without compromising bone marrow edema pattern accuracy

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS10062167B2Estimated local rigid regions from dense deformation in subtraction

Publication Date: 2018.08.28 TOSHIBA MEDICAL SYST CORP

AI summary

A medical image data processing apparatus comprises a registration unit configured to perform a non-rigid registration of a first set of medical image data and a second set of medical image data thereby to determine a deformation field, and a region identification unit configured to identify at least one region for which the deformation field is approximately homogeneous, and, for at least part of at least one identified region, to obtain a substantially homogeneous approximation of the deformation field.