Method of segmenting a three-dimensional structure contained in an object, notably for medical image analysis
US7167172B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Mar 5, 2002 |
| Grant date | Jan 23, 2007 |
| Priority date | — |
| Expiry date | Oct 20, 2023 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/30004
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
The invention relates to a method of segmenting a three-dimensional structure, contained in an object, from one or more two-dimensional images which represent a slice of the object. The method utilizes a deformable model whose surface is formed by a network of meshes which connect network points on the surface of the model to one another. First there are determined the meshes which intersect at least one image and a point on the surface of the structure to be segmented is searched along a search line which traverses the mesh and extends in the image. Subsequently, the position of the network points of the model is calculated anew. These steps are repeated a number of times and the model ultimately obtained, that is, after several deformations, is considered to be the segmentation of the three-dimensional structure from the two-dimensional images.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.