Sciweavers

Share
ICPR
2006
IEEE

Automatic Hip Bone Segmentation Using Non-Rigid Registration

10 years 2 months ago
Automatic Hip Bone Segmentation Using Non-Rigid Registration
This paper presents a method for automatic segmentation of bone from volumetric computed tomography (CT) data. Due to osteoporosis, which degenerates the bone density and hence decreases the intensity of the bone in the CT dataset, it is not possible to use conventional thresholding techniques to handle the segmentation. Furthermore we want to use prior knowledge about shapes and relations of the bones in the area of interest to be able to e.g. separate adjoining bones from each other. The method we suggest is the morphon algorithm [4]. This is a non-rigid registration technique where an 2D or 3D image is iteratively deformed to match the corresponding structure in a target image. The method uses difference in local quadrature phase and certainty measures to estimate the deformations.
Hans Knutsson, Johanna Pettersson, Magnus Borga
Added 09 Nov 2009
Updated 09 Nov 2009
Type Conference
Year 2006
Where ICPR
Authors Hans Knutsson, Johanna Pettersson, Magnus Borga
Comments (0)
books