Abstract
Benign Prostatic Hyperplasia has two histologically distinct components. It is important to identify which of the two is predominant, as this influences treatment. The two forms can be distinguished radiologically on MR images using a number of visual and semi-quantitative cues. In this paper we investigate the use of texture features based on co-occurrence histograms to provide a quantitative assessment of the tissue type. We show that appropriate texture measures associate strongly with radiological classification based on published criteria.
Original language | English |
---|---|
Title of host publication | Proceedings of Medical Image Understanding and Analysis |
Editors | James Graham, Neil Thacker, Timothy Cootes |
Place of Publication | Manchester, UK |
Publisher | BMVA Press |
Pages | 231-235 |
Number of pages | 5 |
ISBN (Print) | 1 901727 31 6 |
Publication status | Published - Jul 2006 |
Event | Medical Image Understanding and Analysis - University of Manchester Duration: 4 Jul 2006 → 5 Jul 2006 |
Conference
Conference | Medical Image Understanding and Analysis |
---|---|
City | University of Manchester |
Period | 4/07/06 → 5/07/06 |