A local probabilistic prior-based active contour model for brain MR image segmentation

Authors:
Jundong Liu;Charles Smith;Hima Chebrolu
Affiliations:
School of Electrical Engineering and Computer Science, Ohio University, Athens, OH;Department of Neurology, University of Kentucky, Lexington, KY;Department of Neurology, University of Kentucky, Lexington, KY
Venue:
ACCV'07 Proceedings of the 8th Asian conference on Computer vision - Volume Part I
Year:
2007

Citing 7
Cited 1

Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A level set algorithm for minimizing the Mumford-Shah functional in image processing

VLSM '01 Proceedings of the IEEE Workshop on Variational and Level Set Methods (VLSM'01)
Segmentation of Edge Preserving Gradient Vector Flow: An Approach Toward Automatically Initializing and Splitting of Snakes

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
A Review of Statistical Approaches to Level Set Segmentation: Integrating Color, Texture, Motion and Shape

International Journal of Computer Vision
Snakes, shapes, and gradient vector flow

IEEE Transactions on Image Processing
Active contours without edges

IEEE Transactions on Image Processing
Image segmentation and selective smoothing by using Mumford-Shah model

IEEE Transactions on Image Processing

Global optimization in discretized parameter space for predefined object segmentation

Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a probabilistic prior-based active contour model for segmenting human brain MR images. Our model is formulated with the maximum a posterior (MAP) principle and implemented under the level set framework. Probabilistic atlas for the structure of interest, e.g., cortical gray matter or caudate nucleus, can be seamlessly integrate into the level set evolution procedure to provide crucial guidance in accurately capturing the target. Unlike other region-based active contour models, our solution uses locally varying Gaussians to account for intensity inhomogeneity and local variations existing in many MR images are better handled. Experiments conducted on whole brain as well as caudate segmentation demonstrate the improvement made by our model.