Experimental Analysis of the MRF Algorithm for Segmentation of Noisy Medical Images

Authors

  • Joe Qranfal Simon Fraser University
  • Dorit S. Hochbaum University of California, Berkeley
  • Germain Tanoh Fraser Health

Keywords:

MRF, Convex Optimization, Denoising, Segmentation, Medical Imaging, Pseudo Flow Algorithm

Abstract

We show here that the implementation of the Markov random fields image segmentation algorithm of Hochbaum 2001 works well for the purpose of denoising and segmenting medical images. One of the main contributions here is the ability for a user to manipulate online the image so as to achieve clear delineation of objects of interest in the image. This is made possible by the efficiency of the implementation. Results are presented for images that are generated by Single Photon Emission Computed Tomography and Magnetic Resonance Imaging. The results show that the method presented is effective at denoising medical images as well as segmenting tissue types, organs, lesions, and other features within medical images. We advocate that this method should be considered as part of the medical imaging toolbox.

Downloads

Published

2012-01-03

How to Cite

Qranfal, J., Hochbaum, D. S., & Tanoh, G. (2012). Experimental Analysis of the MRF Algorithm for Segmentation of Noisy Medical Images. Algorithmic Operations Research, 6(2), Pages 79 – 90. Retrieved from https://journals.lib.unb.ca/index.php/AOR/article/view/18234

Issue

Section

Articles