DocumentCode :
3509598
Title :
Combining atlas and active contour for automatic 3D medical image segmentation
Author :
Gao, Yi ; Tannenbaum, Allen
Author_Institution :
Dept. of Biomed. Eng., Georgia Inst. of Technol., Atlanta, GA, USA
fYear :
2011
fDate :
March 30 2011-April 2 2011
Firstpage :
1401
Lastpage :
1404
Abstract :
Atlas based methods and active contours are two families of techniques widely used for the task of 3D medical image segmentation. In this work we present a coupled framework where the two methods are combined together, in order to exploit each´s advantage while avoid their respective drawbacks. Indeed, the atlas based methods lacks the flexibility in locally tuning the segmentation boundary; whereas the active contour has the drawback that the final result heavily depends on the initialization as well as the contour evolution energy functional. Therefore, in the proposed work, the atlas based segmentation provides a probability map, which not only supplies the initial contour position, but also defines the contour evolution energy in an on-line fashion. Afterward, the active contour further converges to the desired object boundary. Finally, the method is tested on various 3D medical images to demonstrate its robustness as well as accuracy.
Keywords :
biomedical MRI; data analysis; image segmentation; learning (artificial intelligence); medical image processing; statistical distributions; 3D medical image segmentation; MRI data sets; active contours; atlas based methods; contour evolution energy; probability map; shape learning; Active contours; Biomedical imaging; Image segmentation; Robustness; Shape; Three dimensional displays; Training; Active contour segmentation; Atlas based segmentation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical Imaging: From Nano to Macro, 2011 IEEE International Symposium on
Conference_Location :
Chicago, IL
ISSN :
1945-7928
Print_ISBN :
978-1-4244-4127-3
Electronic_ISBN :
1945-7928
Type :
conf
DOI :
10.1109/ISBI.2011.5872662
Filename :
5872662
Link To Document :
بازگشت