Method and system for navigating, segmenting, and extracting a three-dimensional image
AssigneeGeneral Electric Company
InventorsZhongmin Lin, Gopal Avinash, Patrick Michael Virtue
Life of this patent
Estimating the term…
Abstract
A method for extracting a three-dimensional (3D) volume of interest from a three-dimensional (3D) image dataset includes accessing a 3D image dataset that includes a plurality of image slices, enclosing a 3D volume of interest in the 3D image dataset using a 3D mesh, automatically extracting the 3D volume of interest based on the 3D mesh, and generating a 3D image of the extracted 3D volume of interest. A computer and a non-transitory computer readable medium are also described herein.
Claims20
A method for extracting a three-dimensional (3D) volume of interest from a three-dimensional (3D) image dataset, said method comprising:
- accessing a 3D image dataset that includes a plurality of image slices;
- enclosing a 3D volume of interest in the 3D image dataset using a 3D mesh, wherein the 3D mesh comprises a plurality of polygons;
- projecting the polygons onto two-dimensional (2D) planes;
- identifying, for each polygon, a set of 2D grid points that are bounded by the respective polygon;
- automatically extracting the 3D volume of interest based on the 3D mesh;
- and
- generating a 3D image of the extracted 3D volume of interest.
A system for extracting a three-dimensional (3D) volume of interest from a three-dimensional (3D) image, said computer comprising:
- a user interface;
- and
- a processor coupled to the user interface, the processor being configured to:
- access a 3D image dataset that includes a plurality of image slices;
- enclose a 3Dvolume of interest in the 3D image dataset using a 3D mesh, wherein the 3D mesh comprises a plurality of polygons;
- project the polygons onto two-dimensional (2D) planes;
- identify, for each polygon, a set of 2D grid points that are bounded by the respective polygon;
- automatically extract the 3D volume of interest based on the 3D mesh;
- and
- generate a 3D image of the extracted 3D volume of interest.
A non-transitory computer readable medium programmed to instruct a computer to:
- access a 3D image dataset that includes a plurality of image slices;
- enclose a 3D volume of interest in the 3D image dataset using a 3D mesh, wherein the 3D mesh comprises a plurality of polygons;
- project the polygons onto two-dimensional (2D) planes;
- identify, for each polygon, a set of 2D grid points that are bounded by the respective polygon;
- automatically extract the 3D volume of interest based on the 3D mesh;
- and
- generate a 3D image of the extracted 3D volume of interest.
Dependencies are inferred from the claim text.