DocumentCode :
406567
Title :
Fast reconstruction of volumetric models of anatomical structures
Author :
Archip, Neculai ; Rohling, Robert
Author_Institution :
Dept. of Electr. & Comput. Eng., British Columbia Univ., BC, Canada
Volume :
1
fYear :
2003
fDate :
17-21 Sept. 2003
Firstpage :
568
Abstract :
This paper reports a fast and robust method for building 3D volumetric models of anatomical structures. The proposed reconstruction solution is image modality independent. As input, the method requires a set of contours obtained after organ segmentation. It is not necessary for contours to lie on parallel slices. The contours are simplified using a combination of smoothing techniques and B-splines. A 3D triangulation of the object convex hull is generated, based on a 3D Delaunay triangulation algorithm. A sculpting process of the non-object tetrahedra is performed, based on the spatial information obtained in the segmentation process. The surface, represented as a set of triangles, and the volume, represented as a set of tetrahedral, are obtained. To validate the algorithm, geometrical measurements of the generated model are computed, and compared with the original object. Currently, the method is tested on CT and MR images. At this stage, 2 seconds are needed on a standard workstation to build a skull model with approximately 120,000 tetrahedrals.
Keywords :
biomedical MRI; computerised tomography; image reconstruction; image segmentation; medical image processing; mesh generation; splines (mathematics); 3D Delaunay triangulation algorithm; 3D triangulation; B-splines; CT images; MR images; anatomical structures; fast reconstruction; image modality; nonobject tetrahedra; object convex hull; organ segmentation; sculpting process; skull model; smoothing techniques; volumetric models; Anatomical structure; Buildings; Computed tomography; Image reconstruction; Image segmentation; Robustness; Smoothing methods; Solid modeling; Spline; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society, 2003. Proceedings of the 25th Annual International Conference of the IEEE
ISSN :
1094-687X
Print_ISBN :
0-7803-7789-3
Type :
conf
DOI :
10.1109/IEMBS.2003.1279808
Filename :
1279808
Link To Document :
بازگشت