A Robust MR Image Segmentation Technique Using Spatial Information and Principle Component Analysis
Automated segmentation of MR images is a difficult problem due to the complexity of the images. Up to now, several approaches have been proposed based on spectral characteristics of MR images, but they are sensitive to the noise contained in the MR images. In this paper, we propose a robust method for noisy MR image segmentation. We use region-based features for a robust segmentation and use principle component analysis (PCA) to reduce the large dimensionality of feature space. Experimental results show that the proposed method is very tolerant to the noise and the segmentation performance is significantly improved.
KeywordsFalse Negative Rate Principle Component Analysis Target Class Segmentation Performance Proton Density Image
Unable to display preview. Download preview PDF.
- 5.Warfield, S.K., Kaus, M., Jolesz, F.A., Kikinis, R.: Adaptive, Template Moderated, Spatially Varying Statistical Classification. 4, 43–55 (2000)Google Scholar
- 7.Mohamed, N.A., Aly, A.F.: Volume Segmentation of CT/MRI Images Using Multiscale Features, Self-organizing Principal Component Analysis, and Self-organizing Feature Map. In: Proc. of Int. Conf. on Artificial Neural Networks, Houston (1997)Google Scholar
- 10.Yoon, O.Y., Kwak, D.M., Kim, B.S., et al.: Automated Segmentation of MR Brain Image Using 3-Dimensional Clustering. IEICE Trains. Inf. & Syst. E85-D, 773–781 (2002)Google Scholar
- 11.Jolliffe, I.T.: Principal Component Analysis. Springer, New York (1986)Google Scholar