Automated Bone Marrow Segmentation in 3D CT Imaging

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for identifying bone marrow in CT images are manual, time-consuming, and prone to inaccuracies due to radiation exposure, as they require tracing regions in full-body image slices, which is tedious and inefficient.

Innovation Solution

A method using 3D CT volume data sets to identify voxels with Hounsfield Unit values below a bone threshold, segmenting non-bone regions, and expanding regions to distinguish bone marrow from bone, allowing for automatic segmentation of bone marrow without relying on image continuity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual tracing of bone marrow regions is performed on full body image slices, then bone marrow identification can be achieved, but the process becomes very tedious and time-consuming

Engineering Contradiction:

Improvebone marrow identification accuracyVSAvoidtime for manual tracing

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical tracing with an automated computer-based system that uses Hounsfield Unit thresholding to identify bone marrow regions. The system automatically segments bone marrow by detecting voxels with HU values below a bone threshold, eliminating the need for manual tracing while maintaining identification accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent uses Hounsfield Unit parameter thresholding to automatically distinguish bone marrow from other tissues. By setting a specific HU threshold value, the system can automatically identify bone marrow regions without manual intervention, transforming the identification process from time-consuming manual tracing to rapid automated parameter-based segmentation.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If manual tracing methods are used to identify bone marrow regions, then bone marrow mass can be measured, but the process is prone to inaccuracies due to radiation exposure risks

Engineering Contradiction:

Improvebone marrow mass measurementVSAvoidradiation exposure risk

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The automated segmentation system replaces manual tracing operations that require repeated patient positioning and imaging. By performing all segmentation operations through computer algorithms on existing CT data, the system eliminates the need for additional radiation exposure that would occur during manual verification and adjustment procedures.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If automatic segmentation based on Hounsfield Unit thresholding is implemented, then processing speed and reproducibility improve, but differentiation between bone and bone marrow requires additional region expansion steps

Engineering Contradiction:

Improvesegmentation processing speedVSAvoidsegmentation algorithm complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent divides the segmentation process into distinct stages: initial identification of non-bone voxels using HU thresholding, region expansion to capture complete bone marrow spaces, and final differentiation of bone marrow from bone. This segmented approach allows each step to be optimized independently, maintaining processing speed while systematically handling the complexity of tissue differentiation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary identification of all non-bone voxels using HU thresholding before conducting region expansion. This preliminary action creates a starting set of voxels that are then expanded to ensure complete capture of bone marrow regions, simplifying the overall differentiation process by separating identification from boundary definition.

Inventive Principle:

Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enables faster and more reproducible automatic segmentation of bone marrow, reducing radiation exposure risks and improving the accuracy of bone marrow identification in CT images.

Implementation Method 1

a gantry supporting an x-ray source and a detector array for rotatable operation to scan a patient to acquire a three-dimensional (3D) image data set

Methodology Applied

Scientific EffectX-ray attenuation: X-Ray

Implementation Method 2

an image reconstructor configured to reconstruct a 3D image of the patient using the 3D image data set

Methodology Applied

Scientific EffectImage reconstruction: Image Processing

Data Source

PatentUS9811903B2Systems and methods for identifying bone marrow in medical images

Publication Date: 2017.11.07 GE PRECISION HEALTHCARE LLC

AI summary

Systems and method for identifying bone marrow in medical images are provided. A method includes obtaining a three-dimensional (3D) computed tomography (CT) volume data set corresponding to an imaged volume and identifying voxels in the 3D CT volume data set having a Hounsfield Unit (HU) value below a bone threshold. The voxels are identified without using image continuity. The method further includes marking the identified voxels as non-bone voxels, determining definite tissue voxels based on the identified non-bone voxels and expanding a region defined by the definite tissue voxels. The method also includes segmenting the expanded region to identify bone voxels and bone marrow voxels and identifying bone marrow as voxels that are not the bone voxels.