The spine bone consists of the vertebral body (VB) and processess shown in the figure. The VB consists of trabecular and cortical bones. The main objective of our algorithm is the segmentation of the vertebral body of the spine bone. In this project, we are primarily interested in volumetric computed tomography (CT) images of the vertebral bones of spine column with a particular focus on the lumbar spine. The primary goal of the proposed work is in the field of spine densitometry where bone mineral density (BMD) measurements and fracture analysis are restricted to the vertebral bodies (see the figure for regions of spine bone [*]).
To assess the accuracy of our segmentation experiment, we tested it using 30 clinical data sets, as well as, the phantom (ESP). The real data sets were scanned at 120kV and 2.5mm slice thickness. The ESP was scanned at 120kV and 1.00mm slice thickness. All algorithms are run on a PC 3Ghz AMD Athlon 64 X2 Dual, and 3GB RAM. All implementations are in C++.
Research Team
Aly A. Farag
CVIP Director
Ben Arnold
Image Analysis, Inc., Director
Melih S. Aslan
Research Assistant
Dongqing Chen
Post Doc
dqchen@cvip.louisville.edu
Asem Ali
Post Doc
Rachid Fahmi
Alumni
Mike Miller
System Integration
Methods
Publications
Melih S. Aslan, Asem Ali, Ben Arnold, Rachid Fahmi, Aly A. Farag, and Ping Xiang, Segmentation of trabecular bones from vertebral bodies in volumetric CT spine images, Proc. of IEEE International Conference on Image Processing (ICIP09),Cairo, Egypt, November 7-11, Accepted to appear.
Melih S. Aslan, Asem Ali, Ham Rara, Ben Arnold , Aly A. Farag, Rachid Fahmi, and Ping Xiang , A Novel 3D Segmentation of Vertebral Bones from Volumetric CT Images Using Graph Cuts, 5th International Symposium on Visual Computing (ISVC-09) , Las Vegas, Neveda, Nov 30-Dec 2, 2009, Accepted to appear.
Acknowledgement / Sponsors
This project is funded by Image Analysis, Inc. in Columbia, KY.
[*]The spine figure is adopted from http://www.back.com.
CVIP Lab 