Multi-modality vertebra recognition

Patent metadata

Field	Value
Publication number	US-9401020-B1
Application number	US-201514701769-A
Country	US
Kind code	B1
Filing date	May 1, 2015
Priority date	May 1, 2015
Publication date	Jul 26, 2016
Grant date	Jul 26, 2016

Abstract

Systems, methods and computer program products to perform multi-modality vertebrae recognition are provided. Aspects of the present disclosure disclosed and described herein enable a multi-modality vertebrae recognition engine that recognizes vertebrae by using a three-stage recognition approach: landmark detection, global shape registration, and local pose adjustment. These stages cover the matching from local to global spine structures. The three stages are implemented by three modules in the hierarchical deformable model, the local appearance module, global geometry model, and local geometry model. According to one aspect of the present disclosure, the overall workflow can be understood as a three-stage top-down registration. The goal of the registration is: 1) to align the global shape of the spine model, and 2) to align the vertebrae poses with the local image structures around identified landmarks.

First claim

What is claimed is:

1. A computer-implemented method to recognize one or more anatomic structures in digital images, the method comprising:
loading a pre-computed 3D spine mesh model with reference data comprising prior geometric measurements of the human spine;
receiving image data comprising one or more images of at least one anatomic structure;
generating, using a processor, one or more planar vertebra appearance templates using said one or more images;
identifying vertebrae landmarks using image feature matching between said one or more images and said one or more planar vertebra appearance templates;
computing, using a processor, the spine shape by aligning said vertebrae landmarks with said pre-computed 3D spine mesh model;
identifying the orientation and dimension of said one or more anatomic structures using a triplanar model and group-wise registration;
computing, using a processor, a 3D spine model using a hierarchical deformable model, said planar vertebra templates, said vertebra landmarks, and said spine shape; and
display said computed 3D model of the spine on said computer.

2. The computer-implemented method of claim 1 , wherein said computed 3D spine model comprises:
identifying local and global spine structures;
identifying spine shape, and
extracting vertebra information comprising vertebra locations, labels and poses.

3. The computer-implemented method of claim 1 , wherein generating said 3D spine model comprises:
using images comprising at least one anatomic structure, from at least one anatomic plane, from at least one imaging modality.

4. The computer-implemented method of claim 2 , wherein the said extracted vertebra information and said identified spine shape are used for diagnostic purposes.

5. The computer-implemented method of claim 1 , wherein the dimensions and orientations of local vertebra parts can be adjusted by a human user.

6. A computer storage device including program instructions for execution by a computing device to perform:
loading a pre-computed 3D spine mesh model with reference data comprising prior geometric measurements of the human spine;
receiving image data comprising one or more images of at least one anatomic structure;
generating, using a processor, one or more planar vertebra appearance templates using said one or more images;
identifying vertebrae landmarks using image feature matching between said one or more images and said one or more planar vertebra appearance templates;
computing, using a processor, the spine shape by aligning said vertebrae landmarks with said pre-computed 3D spine mesh model;
identifying the orientation and dimension of said one or more anatomic structures using a triplanar model and group-wise registration;
computing, using a processor, a 3D spine model using a hierarchical deformable model, said planar vertebra templates, said vertebra landmarks, and said spine shape; and
display said computed 3D model of the spine on said computer.

7. The computer storage device of claim 6 , further including program instructions for execution by said computing device to perform:
using said 3D spine model to:
identify local and global spine structures;
identify spine shape, and
extract vertebra information comprising vertebra locations, labels and poses.

8. The computer storage device of claim 6 , further including program instructions for execution by said computing device to perform:
generating the 3D spine model comprises using images comprising at least one anatomic structure, from at least one anatomic plane, from at least one imaging modality.

9. The computer storage device of claim 7 , further including program instructions for execution by said computing device to perform:
extracting vertebra information and said identified spine shape to be used for diagnostic purposes.

10. The computer storage device of claim 6 , further including program instructions for execution by said computing device to perform:
allowing a human user to adjust the dimensions and orientations of local vertebra parts.

11. A system comprising a processor, the processor configured to execute computer program instructions to:
load a pre-computed 3D spine mesh model with reference data comprising prior geometric measurements of the human spine;
receive image data comprising one or more images of at least one anatomic structure;
generate, using a processor, one or more planar vertebra appearance templates using said one or more images;
identify vertebrae landmarks using image feature matching between said one or more images and said one or more planar vertebra appearance templates;
compute, using a processor, the spine shape by aligning said vertebrae landmarks with said pre-computed 3D spine mesh model;
identify the orientation and dimension of said one or more anatomic structures using a triplanar model and group-wise registration;
compute, using a processor, a 3D spine model using a hierarchical deformable model, said planar vertebra templates, said vertebra landmarks, and said spine shape; and
display said computed 3D model of the spine on said computer.

12. The system of claim 11 , wherein the system further comprises:
identifying local and global spine structures;
identifying spine shape, and
extracting vertebra information comprising vertebra locations, labels and poses.

13. The system of claim 11 , wherein the system further comprises:
generating said 3D spine model using images comprising at least one anatomic structure, from at least one anatomic plane, from at least one imaging modality.

14. The system of claim 12 , wherein the system further comprises:
extracting said vertebra information and said identified spine shape are used for diagnostic purposes.

15. The system of claim 11 , wherein the system further comprises:
allowing a human user to adjust the dimensions and orientations of local vertebra parts.

Assignees

• Gen Electric
• London Health Sci Ct Res Inc
• London Health Science Centre Res Inc

Inventors

• Li Shuo
• Cai Yungliang
• Osman Said
• Landis Mark
• Sharma Manas

Classifications

• G06T7/74 Primary

  involving reference images or patches · CPC title

• G06V10/754

  involving a deformation of the sample pattern or of the reference pattern; Elastic matching · CPC title

• G06T7/0012 Primary

  Biomedical image inspection · CPC title

• G06K2209/05

  Physics · mapped topic

• G06T2207/30012

  Spine; Backbone · CPC title