Title:	BRAIN TUMOR SEGMENTATION FROM MRI: A COMPARATIVE STUDY
Author(s):	Simon L. Cichosz, Steffen Vangsgaard, Alex S. Jørgensen, Kasper E. Kannik, Elena Steffensen, Simon F. Eskildsen
ISBN:	978-972-8939-22-9
Editors:	Yingcai Xiao, Tomaz Amon and Piet Kommers
Year:	2010
Edition:	Single
Keywords:	Segmentation, MRI, brain tumor, active contour
Type:	Short Paper
First Page:	401
Last Page:	406
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Identification of tumor extent and proliferation is an important step in using MRI techniques for tumor grading purposes and treatment planning. Today tumor segmentation is a manual task done by trained operators. The segmentation is both time consuming and subject to inter- and intraoperator variability which may interfere with an accurate diagnosis and treatment evaluation. In this paper five semi-automatic methods for tumor segmentation are presented and tested on clinical MRI data. These methods include voxel/pixel and contour based segmentations and are compared to manual delineations. The five algorithms showed varying segmentation accuracy and reliability. The results suggest that improvements are needed to meet the requirements for use in clinical practice.
	Go Back