Medical image segmentation using cooccurrence matrix based texture features calculated on weighted region

Authors:
Ludvik Tesar;Daniel Smutek;Akinobu Shimizu;Hidefume Kobatake
Affiliations:
Tokyo University of Agriculture and Technology, Japan and Institute of Inf. Theory and Automation, Czech Academy of Science;Charles University, Prague, Czech Republic;Tokyo University of Agriculture and Technology, Japan;Tokyo University of Agriculture and Technology, Japan
Venue:
ACST'07 Proceedings of the third conference on IASTED International Conference: Advances in Computer Science and Technology
Year:
2007

Citing 3
Cited 2

Computer and Robot Vision

Computer and Robot Vision
Systematic Construction of Texture Features for Hashimoto's Lymphocytic Thyroiditis Recognition from Sonographic Images

AIME '01 Proceedings of the 8th Conference on AI in Medicine in Europe: Artificial Intelligence Medicine
3D extension of Haralick texture features for medical image analysis

SPPRA '07 Proceedings of the Fourth IASTED International Conference on Signal Processing, Pattern Recognition, and Applications

Active Contours Driven by Supervised Binary Classifiers for Texture Segmentation

ISVC '08 Proceedings of the 4th International Symposium on Advances in Visual Computing
A novel T-CAD framework to support medical image analysis and reconstruction

ICIAP'11 Proceedings of the 16th international conference on Image analysis and processing - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

An improvement of texture based 2D or 3D image segmentation method is proposed, aimed at medical image analysis. Segmentation of organs or disease diagnosing is the targetted problem. Features based on cooccurrence matrix are studied. A new approach to local cooccurrence matrix calculation is proposed in this paper. Separate cooccurrence matrix is calculated for each pixel (in 2D case) or vowel (in 3D case), based on pixels (voxels) around the original pixel (voxel). The new approach is in the way of calculating the cooccurrence matrix. As opposed to classical definition where cooccurrence matrix values are counts of relative frequencies of cooccurrences over defined region, we propose for each cooccurrence, to be weighted by the function of distance from the original pixel (voxel) of interest. Compared to calculating the cooccurrence texture features over square (or cubic) region around a pixel (voxel), the proposed approach makes cooccurrence matrix and texture feature more focused, therefore it can be used to search for smaller regions with different texture properties (like tumours). A set of abdominal CT images is used for evaluation of the proposed approach and comparison with older cubic region based approach. The actual cooccurrence features used were some of those proposed by Haralick.