Lesion boundary segmentation using level set methods

Massey, Elizabeth and Lowell, James and Hunter, Andrew and Steele, David (2009) Lesion boundary segmentation using level set methods. In: VisApp - Proceedings of the Fourth International Conference on Computer Vision Theory and Applications - Volume 1, February 5-8, 2009, Lisbon, Portugal.

Documents
Lesion boundary segmentation using level set methods
This paper addresses the issue of accurate lesion segmentation in retinal imagery, using level set methods and a novel stopping mechanism - an elementary features scheme. Specifically, the curve propagation is guided by a gradient map built using a combination of histogram equalization and robust statistics. The stopping mechanism uses elementary features gathered as the curve deforms over time, and then using a lesionness measure, defined herein, ’looks back in time’ to find the point at which the curve best fits the real object. We implement the level set using a fast upwind scheme and compare the proposed method against five other segmentation algorithms performed on 50 randomly selected images of exudates with a database of clinician marked-up boundaries as ground truth.
[img]
[Download]
[img]
Preview
PDF (Requires Adobe Acrobat Reader)
LBFitLS-VisApp09.pdf

252kB

Official URL: http://www.pubzone.org/dblp/conf/visapp/MasseyLHS0...

Abstract

This paper addresses the issue of accurate lesion segmentation in retinal imagery, using level set methods and
a novel stopping mechanism - an elementary features scheme. Specifically, the curve propagation is guided by a gradient map built using a combination of histogram equalization and robust statistics. The stopping mechanism uses elementary features gathered as the curve deforms over time, and then using a lesionness measure, defined herein, ’looks back in time’ to find the point at which the curve best fits the real object. We implement the level set using a fast upwind scheme and compare the proposed method against five other
segmentation algorithms performed on 50 randomly selected images of exudates with a database of clinician
marked-up boundaries as ground truth.

Item Type:Conference or Workshop Item (Paper)
Additional Information:This paper addresses the issue of accurate lesion segmentation in retinal imagery, using level set methods and a novel stopping mechanism - an elementary features scheme. Specifically, the curve propagation is guided by a gradient map built using a combination of histogram equalization and robust statistics. The stopping mechanism uses elementary features gathered as the curve deforms over time, and then using a lesionness measure, defined herein, ’looks back in time’ to find the point at which the curve best fits the real object. We implement the level set using a fast upwind scheme and compare the proposed method against five other segmentation algorithms performed on 50 randomly selected images of exudates with a database of clinician marked-up boundaries as ground truth.
Keywords:computer vision, retinal lesion segmentation, segmentation, level set methods
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
Divisions:College of Science > School of Computer Science
ID Code:1658
Deposited By:INVALID USER
Deposited On:15 Oct 2008 18:00
Last Modified:13 Mar 2013 08:30

Repository Staff Only: item control page