Compare different spatial based fuzzy-C_mean (FCM) extensions for MRI image segmentation

Author

Balafar, M.A. ; Ramli, A.R. ; Mashohor, S. ; Farzan, A.

Author_Institution

Dept of Comput. & Commun. Syst., Univ. Putra Malaysia, Serdang, Malaysia

Volume

5

fYear

2010

fDate

26-28 Feb. 2010

Firstpage

609

Lastpage

611

Abstract

FCM does not use spatial information in clustering process. Therefore, it is not robust against noise and other imaging artefacts. In order to incorporate spatial information, an extension for FCM (FCM_S) is proposed which allows pixel to be labelled by influence of its neighbourhood labels. FCM_S is time-consuming. To over come this problem, FCM_S1 is introduced, which is faster. Then, FCM_EN and FGFCM are proposed which are faster than previous methods. Four spatial based extensions are simulated for FCM: FCM_S, FCM_S1, FCM_EN and FGFCM. In order to compare their quality, they are applied to simulated brain MRI images and similarity index is used to compare their quality quantitatively.

Keywords

biomedical MRI; brain; fuzzy set theory; image segmentation; FCM extension; FCM_EN; FCM_S1; FGFCM; MRI image segmentation; brain segmentation; spatial based fuzzy-C_mean extension; Biomedical imaging; Brain modeling; Clustering algorithms; Computed tomography; Image segmentation; Magnetic resonance imaging; Medical diagnostic imaging; Noise robustness; Systems engineering and theory; Unsupervised learning; Brain segmentation; FCM;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Automation Engineering (ICCAE), 2010 The 2nd International Conference on

Conference_Location

Singapore

Print_ISBN

978-1-4244-5585-0

Electronic_ISBN

978-1-4244-5586-7

Type

conf

DOI

10.1109/ICCAE.2010.5451302

Filename

5451302