3D joint Markov-Gibbs model for segmenting the blood vessels from MRA

Authors:
Ayman El-Baz;Georgy Gimel'farb;Vedant Kumar;Robert Falk;Mohamed Abo El-Ghar
Affiliations:
BioImaging Laboratory, Bioengineering Department, University of Louisville, Louisville, KY;Computer Science Department, University of Auckland, New Zealand;BioImaging Laboratory, Bioengineering Department, University of Louisville, Louisville, KY;Director Medical Imaging Division, Jewish Hospital, Louisville, Kentucky;Urology and Nephrology Department, University of Mansoura, Mansoura, Egypt
Venue:
ISBI'09 Proceedings of the Sixth IEEE international conference on Symposium on Biomedical Imaging: From Nano to Macro
Year:
2009

Citing 4
Cited 0

Geodesic Active Contours

International Journal of Computer Vision
Medical image segmentation using topologically adaptable surfaces

CVRMed-MRCAS '97 Proceedings of the First Joint Conference on Computer Vision, Virtual Reality and Robotics in Medicine and Medial Robotics and Computer-Assisted Surgery
Fast Extraction of Tubular and Tree 3D Surfaces with Front Propagation Methods

ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 1 - Volume 1
Snakes, shapes, and gradient vector flow

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

New techniques for more accurate segmentation of a 3D cerebrovascular system from time-of-flight (TOF) magnetic resonance angiography (MRA) data are proposed. In this paper, we describe TOF-MRA images and desired maps of regions (blood vessels and the other brain tissues) by a joint Markov-Gibbs random field model (MGRF) of independent image signals and interdependent region labels but focus on most accurate model identification. To better specify region borders, each empirical distribution of signals is precisely approximated by a Linear Combination of Discrete Gaussians (LCDG) with positive and negative components. We modify a conventional Expectation-Maximization (EM) algorithm to deal with the LCDG and develop a sequential EM-based technique to get an initial LCDG-approximation for the modified EM algorithm. The initial segmentation based on the LCDG-models is then iteratively refined using a GMRF model with analytically estimated potentials. To validate the accuracy of our algorithm, a special 3D geometrical phantom motivated by statistical analysis of the MRA-TOF data is designed. Experiments with both the phantom and 50 real data sets confirm high accuracy of the proposed approach.