Tissue classification using image intensities and anatomical positions

Patent metadata

Field	Value
Publication number	US-11270436-B2
Application number	US-201916961706-A
Country	US
Kind code	B2
Filing date	Jan 9, 2019
Priority date	Jan 16, 2018
Publication date	Mar 8, 2022
Grant date	Mar 8, 2022

Abstract

The invention relates to a medical image data processing system (101) for image segmentation. The medical image data processing system (101) comprises a machine learning framework trained to receive an anatomical position of a voxel and to provide a tissue type classification. An execution of machine executable instructions by a processor (130) of the medical image data processing system (101) causes the processor (130) to control the medical image data processing system (101) to: —receive medical image data (140) comprising an anatomical structure of interest, —fit an anatomical frame of reference (302, 402) to the medical image data (140) using model-based segmentation, —classify tissue types represented by voxels of the medical image data (140) using the machine learning framework, wherein anatomical positions of the voxels with respect to the anatomical frame of reference (302, 402) are used as the input to the machine learning framework.

First claim

The invention claimed is:

1. A medical image data processing system, the medical image data processing system comprising:
a memory storing machine executable instructions and a machine learning framework trained to receive an image intensity and an anatomical position of a voxel as an input and to provide a tissue type classification of the voxel in response as an output,
a processor for controlling the medical image data processing system, wherein execution of the machine executable instructions by the processor causes the processor to control the medical image data processing system to:
receive medical image data comprising an anatomical structure of interest,
fit an anatomical frame of reference to the medical image data using model-based segmentation, wherein a model used for the model-based segmentation comprises an anatomical reference structure in a reference space defined by the anatomical frame of reference, wherein the fitting of the anatomical frame of reference comprises deforming the reference space together with the anatomical reference structure to align the deformed anatomical reference structure with the anatomical structure of interest, wherein the fitting of the anatomical frame of reference provides anatomical positions relative to the deformed anatomical reference structure
classify tissue types represented by voxels of the medical image data using the machine learning framework, wherein each voxel comprises an image intensity and wherein the image intensity and the anatomical positions of the voxels with respect to the deformed anatomical reference structure are used as the input to the machine learning framework.

2. The medical image data processing system of claim 1 , wherein the medical image data comprises magnetic resonance image data.

3. The medical image data processing system of claim 2 , wherein the medical image data processing system further comprises a magnetic resonance imaging system and wherein the magnetic resonance imaging system comprises:
a main magnet for generating a main magnetic field within an imaging zone,
a magnetic field gradient system for generating a spatially dependent gradient magnetic field within the imaging zone,
a radio-frequency antenna system configured for acquiring magnetic resonance data from the imaging zone,
wherein the memory further stores pulse sequence commands, wherein the pulse sequence commands are configured for controlling the magnetic resonance imaging system to acquire the magnetic resonance data from the imaging zone,
wherein the receiving of the medical image data comprises the execution of the machine executable instructions using pulse sequence commands and acquire the medical image data in form of magnetic resonance image data from the imaging zone by the radio-frequency antenna system.

4. The medical image data processing system of claim 2 , wherein the execution of the machine executable instructions further causes the processor to generate a pseudo-CT image using the magnetic resonance image data and the classification result.

5. The medical image data processing system of claim 1 , wherein the model used for the model-based segmentation comprises the anatomical reference structure in form of a surface mesh used for the segmentation.

6. The medical image data processing system of claim 5 , wherein the model comprises the anatomical frame of reference in form of a spatial reference frame of the surface mesh and wherein the model-based segmentation comprises deforming the anatomical frame of reference together with the surface mesh.

7. The medical image data processing system of claim 1 , wherein at least two voxels with identical image intensities are assigned to different classes representing different tissue types based on different anatomical positions of the voxels with respect to the anatomical frame of reference.

8. The medical image data processing system of claim 7 , wherein a first one of the different classes represents bone and a second one of the different classes represents air.

9. The medical image data processing system of claim 1 , wherein the receiving of the medical image data comprises: sending a request for the respective medical image data to a database comprising the medical image data, wherein in response to the request the requested medical image data is received from the database.

10. A method for controlling a medical image data processing system, the medical image data processing system comprising:
a memory storing machine executable instructions and a machine learning framework trained to receive an image intensity and an anatomical position of a voxel as an input and to provide a tissue type classification of the voxel in response as an output,
a processor for controlling the medical image data processing system, wherein execution of the machine executable instructions by the processor causes the processor to control the medical image data processing system to execute the method comprising:
receiving medical image data comprising an anatomical structure of interest,
fitting an anatomical frame of reference to the medical image data using model-based segmentation, wherein a model used for the model-based segmentation comprises an anatomical reference structure in a reference space defined by the anatomical frame of reference, wherein the fitting of the anatomical frame of reference comprises deforming the reference frame together with the anatomical reference structure to align the deformed anatomical reference structure with the anatomical structure of interest, wherein the fitting of the anatomical frame of reference provides anatomical positions relative to the deformed anatomical reference structure,
classifying tissue-types represented by voxels of the medical image data using the machine learning framework, wherein each voxel comprises an image intensity and wherein the anatomical positions of the voxels with respect to the deformed anatomical reference structure are used as the input to the machine learning framework.

11. The method of claim 10 , wherein the medical image data processing system further comprises a magnetic resonance imaging system and wherein the magnetic resonance imaging system comprises:
a main magnet for generating a main magnetic field within an imaging zone,
a magnetic field gradient system for generating a spatially dependent gradient magnetic field within the imaging zone,
a radio-frequency antenna system configured for acquiring magnetic resonance data from the imaging zone,
wherein the memory further stores pulse sequence commands, wherein the pulse sequence commands are configured for controlling the magnetic resonance imaging system to acquire the magnetic resonance data from the imaging zone,
wherein the receiving of the medical image data comprises the execution of the machine executable instructions using pulse sequence commands and acquire the medical image data in form of magnetic resonance image data from the imaging zone by the radio-frequency antenna system.

12. A computer program product for controlling a medical image data processing system comprising machine executable instructions stored on a non-transitory computer readable medium for execution by a processor controlling the medical image processing system, wherein the medical image data processing system comprises a memory storing a machine learning framework trained to receive an image intensity and an anatomical position of a voxel as an input and to provide a tissue type classification of the voxel in response as an output and a processor for controlling the medical image data processing system, wherein execution of the machine executable instructions

Assignees

• Koninklijke Philips Nv

Inventors

• Buerger Christian
• Renisch Steffen

Classifications

• G06T7/12 Primary

  Edge-based segmentation · CPC title

• G06V10/24

  Aligning, centring, orientation detection or correction of the image · CPC title

• G06T7/0014 Primary

  using an image reference approach · CPC title

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title