DocumentCode :
2720508
Title :
Neck localization and geometry quantification of intracranial aneurysms
Author :
Sgouritsa, E. ; Mohamed, A. ; Morsi, H. ; Shaltoni, H. ; Mawad, M.E. ; Kakadiaris, I.A.
Author_Institution :
Dept. of Comput. Sci., Univ. of Houston, Houston, TX, USA
fYear :
2010
fDate :
14-17 April 2010
Firstpage :
1057
Lastpage :
1060
Abstract :
We present an approach for accurate localization of the neck of intracranial aneurysms and quantification of their geometry that is useful for their treatment through endovascular embolization. In particular, we first obtain a vessel segmentation using a topology-preserving level set method and extract the surface of the segmented vessel. We then separate the aneurysm from the parent vessels and localize its neck by formulating the aneurysm segmentation problem as an s-t minimum cut problem. Finally, we estimate clinically relevant geometric parameters of the aneurysm. The results indicate that there is good agreement between the measurements obtained by the proposed approach and two independent manual sets of measurements obtained by two experienced interventional neuroradiologists.
Keywords :
biomedical MRI; biomedical measurement; blood vessels; brain; feature extraction; image segmentation; medical disorders; medical image processing; neurophysiology; endovascular embolization; geometry quantification; intracranial aneurysms; neck localization; s-t minimum cut problem; surface extraction; topology-preserving level set method; vessel segmentation; Aneurysm; Arteries; Biomedical measurements; Computational geometry; Costs; Embolization; Image segmentation; Level set; Neck; Solid modeling; Angiography; endovascular embolization; intracranial aneurysm; neck localization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical Imaging: From Nano to Macro, 2010 IEEE International Symposium on
Conference_Location :
Rotterdam
ISSN :
1945-7928
Print_ISBN :
978-1-4244-4125-9
Electronic_ISBN :
1945-7928
Type :
conf
DOI :
10.1109/ISBI.2010.5490172
Filename :
5490172
Link To Document :
بازگشت