### Abstract

This paper describes a new model-based segmentation technique combining desirable properties or physical models (snakes), shape representation by Fourier parametrization, and modelling of natural shape variability. Flexible parametric shape models are represented by a parameter vector describing the mean contour and by a set of eigenmodes of the parameters characterizing the shape variation. Usually the segmentation process is divided into an initial placement of the mean model and an elastic deformation restricted to the model variability. This, however, leads to a separation of biological variation due to a global similarity transform from small-scale shape changes originating from elastic deformations of the normalized model contours only. The performance can he considerably improved by building shape models normalised with respect to a small set of stable landmarks (AC-PC in our application) and by explaining the remaining variability among a series of images with the model flexibility. This way the image interpretation is solved by a new coarse-to-fine segmentation procedure based on the set of deformation eigenmodes, making a separate initialization step unnecessary. Although straightforward, the extension to 3-D is severely impeded by difficulties arising during the generation of a proper surface parametrization for arbitrary objects with spherical topology. We apply a newly developed surface parametrization which achieves a uniform mapping between object surface and parameter space. The 3-D procedure is demonstrated by segmenting deep structures of the human brain from MR volume data.

Original language | English (US) |
---|---|

Pages (from-to) | 19-34 |

Number of pages | 16 |

Journal | Medical Image Analysis |

Volume | 1 |

Issue number | 1 |

State | Published - 1996 |

### Fingerprint

### Keywords

- 3-D deformable models
- 3-D shape analysis
- Segmentation of multidimensional images
- Statistical analysis of anatomical objects

### ASJC Scopus subject areas

- Computer Graphics and Computer-Aided Design
- Computer Vision and Pattern Recognition
- Biomedical Engineering
- Radiology Nuclear Medicine and imaging
- Medicine (miscellaneous)
- Computer Science (miscellaneous)

### Cite this

*Medical Image Analysis*,

*1*(1), 19-34.

**Segmentation of 2-D and 3-D objects from MRI volume data using constrained elastic deformations of flexible Fourier contour and surface models.** / Székely, Gabor; Kelemen, András; Breehbühler, Christian; Gerig, Guido.

Research output: Contribution to journal › Article

*Medical Image Analysis*, vol. 1, no. 1, pp. 19-34.

}

TY - JOUR

T1 - Segmentation of 2-D and 3-D objects from MRI volume data using constrained elastic deformations of flexible Fourier contour and surface models

AU - Székely, Gabor

AU - Kelemen, András

AU - Breehbühler, Christian

AU - Gerig, Guido

PY - 1996

Y1 - 1996

N2 - This paper describes a new model-based segmentation technique combining desirable properties or physical models (snakes), shape representation by Fourier parametrization, and modelling of natural shape variability. Flexible parametric shape models are represented by a parameter vector describing the mean contour and by a set of eigenmodes of the parameters characterizing the shape variation. Usually the segmentation process is divided into an initial placement of the mean model and an elastic deformation restricted to the model variability. This, however, leads to a separation of biological variation due to a global similarity transform from small-scale shape changes originating from elastic deformations of the normalized model contours only. The performance can he considerably improved by building shape models normalised with respect to a small set of stable landmarks (AC-PC in our application) and by explaining the remaining variability among a series of images with the model flexibility. This way the image interpretation is solved by a new coarse-to-fine segmentation procedure based on the set of deformation eigenmodes, making a separate initialization step unnecessary. Although straightforward, the extension to 3-D is severely impeded by difficulties arising during the generation of a proper surface parametrization for arbitrary objects with spherical topology. We apply a newly developed surface parametrization which achieves a uniform mapping between object surface and parameter space. The 3-D procedure is demonstrated by segmenting deep structures of the human brain from MR volume data.

AB - This paper describes a new model-based segmentation technique combining desirable properties or physical models (snakes), shape representation by Fourier parametrization, and modelling of natural shape variability. Flexible parametric shape models are represented by a parameter vector describing the mean contour and by a set of eigenmodes of the parameters characterizing the shape variation. Usually the segmentation process is divided into an initial placement of the mean model and an elastic deformation restricted to the model variability. This, however, leads to a separation of biological variation due to a global similarity transform from small-scale shape changes originating from elastic deformations of the normalized model contours only. The performance can he considerably improved by building shape models normalised with respect to a small set of stable landmarks (AC-PC in our application) and by explaining the remaining variability among a series of images with the model flexibility. This way the image interpretation is solved by a new coarse-to-fine segmentation procedure based on the set of deformation eigenmodes, making a separate initialization step unnecessary. Although straightforward, the extension to 3-D is severely impeded by difficulties arising during the generation of a proper surface parametrization for arbitrary objects with spherical topology. We apply a newly developed surface parametrization which achieves a uniform mapping between object surface and parameter space. The 3-D procedure is demonstrated by segmenting deep structures of the human brain from MR volume data.

KW - 3-D deformable models

KW - 3-D shape analysis

KW - Segmentation of multidimensional images

KW - Statistical analysis of anatomical objects

UR - http://www.scopus.com/inward/record.url?scp=0030088682&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030088682&partnerID=8YFLogxK

M3 - Article

C2 - 9873919

AN - SCOPUS:0030088682

VL - 1

SP - 19

EP - 34

JO - Medical Image Analysis

JF - Medical Image Analysis

SN - 1361-8415

IS - 1

ER -