A Fuzzy Clustering Technique for Medical Image Segmentation

Author

Tabakov, Martin

Author_Institution

Inst. of Appl. Informatics, Wroclaw Univ. of Technol.

fYear

2006

fDate

7-9 Sept. 2006

Firstpage

118

Lastpage

122

Abstract

The main objective of medical image segmentation is to extract and characterise anatomical structures with respect to some input features or expert knowledge. This paper describes a way of medical image segmentation using an appropriately defined fuzzy clustering method based on a fuzzy similarity relation. The considered relation is defined in terms of the Euclidean metric. A fuzzy similarity relation-based image segmentation algorithm is also introduced. To illustrate the obtained segmentation process some examples of computed tomography imaging are considered. Some results, using the classical fuzzy c-means clustering algorithm are also presented, for a comparison purpose

Keywords

feature extraction; fuzzy set theory; image segmentation; medical image processing; pattern clustering; Euclidean metric; anatomical structure; classical fuzzy c-means clustering; feature extraction; fuzzy similarity relation; medical image segmentation; Anatomical structure; Biomedical imaging; Clustering algorithms; Clustering methods; Computed tomography; Data mining; Fuzzy systems; Image segmentation; Machine learning algorithms; Pixel;

fLanguage

English

Publisher

ieee

Conference_Titel

Evolving Fuzzy Systems, 2006 International Symposium on

Conference_Location

Ambleside

Print_ISBN

0-7803-9718-5

Electronic_ISBN

0-7803-9719-3

Type

conf

DOI

10.1109/ISEFS.2006.251140

Filename

4016704