Analyzing Wavelet and Bidimensional Empirical Mode Decomposition of MRI Segmentation using Fuzzy C-Means Clustering

  • Gulam Sarwar Chuwdhury Department of Computer Science and Engineering, International Islamic University Chittagong, Chittagong 4203
  • Md. Khaliluzzaman Department of Computer Science and Engineering, International Islamic University Chittagong, Chittagong 4203
  • Md. Rashed-Al Mahfuz Department of Computer Science and Engineering, University of Rajshahi, Rajshahi 6205
Keywords: Image segmentation, fuzzy C-means, magnetic resonance imaging, wavelet, BEMD, SNR

Abstract

Image segmentation is a vital step in medical image processing. Magnetic resonance imaging (MRI) is used for brain tissues extraction in white and gray matter. These tissues extraction help in image segmentation applications such as radiotherapy planning, clinical diagnosis, treatment planning. This paper presents utilization of fuzzy C-means (FCM) clustering by using wavelet and bidimensional empirical mode decomposition (BEMD) to improve the quality of noisy MR images. The signal to noise ratio (SNR) value is calculated from FCM clustering data to examine the best segmentation technique. The experiment with synthetic Brain Web images has demonstrated the efficiency and robustness of the appropriate approach in segmenting medical MRI.

Downloads

Download data is not yet available.
Abstract
1428
PDF
651
Published
2016-11-19
How to Cite
Chuwdhury, G., Khaliluzzaman, M., & Mahfuz, M. R.-A. (2016). Analyzing Wavelet and Bidimensional Empirical Mode Decomposition of MRI Segmentation using Fuzzy C-Means Clustering. Rajshahi University Journal of Science and Engineering, 44, 101-112. https://doi.org/10.3329/rujse.v44i0.30395