Title :
Boundary detection in echocardiography using a Split Bregman edge detector and a topology preserving level set approach
Author :
Duggan, N. ; Schaeffer, Hayden ; Le Guyader, Carole ; Jones, E. ; Glavin, M. ; Vese, L.
Author_Institution :
Electr. & Electron. Eng., Nat. Univ. of Ireland, Galway, Ireland
Abstract :
In the current paper a novel approach for echocardiographic segmentation is proposed based on a combination of the Geometric Active Contour Split Bregman (GSB) method with a topology preserving level set method. The proposed method was tested against manual delineations on 20 frames across 2 datasets and achieved an average Hausdorff distance of 4.01 ± 1.06 mm and Mean Absolute distance of 1.62 ± 0.3 mm, which represented an enhanced performance when compared with intensity gradient and region based methods.
Keywords :
echocardiography; edge detection; gradient methods; image segmentation; medical image processing; topology; average Hausdorff distance; boundary detection; echocardiographic segmentation; geometric active contour split Bregman method; intensity gradient based method; intensity region based method; split Bregman edge detector; topology preserving level set method; Active contours; Detectors; Echocardiography; Image edge detection; Image segmentation; Level set; Topology; Echocardiography; Geometric Split Bregman; Segmentation; Topology Preserving Level Set Method;
Conference_Titel :
Biomedical Imaging (ISBI), 2013 IEEE 10th International Symposium on
Conference_Location :
San Francisco, CA
Print_ISBN :
978-1-4673-6456-0
DOI :
10.1109/ISBI.2013.6556415