Picture of person typing on laptop with programming code visible on the laptop screen

World class computing and information science research at Strathclyde...

The Strathprints institutional repository is a digital archive of University of Strathclyde's Open Access research outputs. Strathprints provides access to thousands of Open Access research papers by University of Strathclyde researchers, including by researchers from the Department of Computer & Information Sciences involved in mathematically structured programming, similarity and metric search, computer security, software systems, combinatronics and digital health.

The Department also includes the iSchool Research Group, which performs leading research into socio-technical phenomena and topics such as information retrieval and information seeking behaviour.

Explore

Semi-automatic segmentation of tongue tumors from magnetic resonance imaging

Doshi, Trushali and Soraghan, John and Petropoulakis, Lykourgos and Gross, Derek and MacKenzie, Kenneth (2013) Semi-automatic segmentation of tongue tumors from magnetic resonance imaging. In: 2013 20th International Conference on Systems, Signals and Image Processing (IWSSIP). IEEE, pp. 143-146. ISBN 9781479909414

Full text not available in this repository. Request a copy from the Strathclyde author

Abstract

Radiation therapy is one of the most effective modalities for treatment of tongue cancer. In order to optimize radiation dose to the tumor region, it is necessary to segment the tumor from normal region. This paper presents a new semiautomatic algorithm that is demonstrated to be able to segment tongue tumor from gadolinium-enhanced T1-weighted magnetic resonance imaging (MRI) to support radiation planning. This algorithm takes sequential MRI slices with visible tongue tumor. The Tumor's region from each slice is segmented using three steps (i) preprocessing, (ii) initialization and (iii) localized region-based level set segmentation. The segmentation results obtained from proposed algorithm are compared with manual segmentation from clinical expert. Results from 9 MRI slices show that there is a good overlap between semi-automatic and manual segmentation results with dice similarity coefficient (DSC) of 0.87±0.05.