Paper Details: Downloads: 505
Serial Number: P1160634001
Title: Discrete Hidden Markov Tree Modelling of Ranklet Transform for Mass Classification in Mammograms
Authors: Abdalla S. Ahmed, Reda A. El-Khoribi, Lubna Fekry
Abstract: In this paper, a new model for mass classification of suspected regions in mammograms is presented. The system uses the quantized ranklet transform of the suspected region as a feature and the discrete data hidden Markov tree as a classifier. The system is then used to train two classes of mammograms (normal and abnormal). Results of testing the system with a set of images shows superior performance in detecting cancerous regions of pre-segmented images with respected to other related systems.
Keywords: Mammograms, classification, hidden Markov tree, ranklet transform.
Journal/Conference: International Journal of Bioinformatics and Medical Engineering
Volume: 6
Issue: 1
Submission Date: 7/1/2006 12:00:00 AM
Review Date: 9/1/2006 12:00:00 AM
Publishing Date: 12/1/2006 12:00:00 AM
Article Downloads: 505
Download:

Facebook