DocumentCode :
2087045
Title :
A algorithm of medical image segmentation based on active contour model
Author :
Li, Haiyun ; Chen, Xiang
Author_Institution :
Capital Med. Univ., Beijing
fYear :
2007
fDate :
23-27 May 2007
Firstpage :
774
Lastpage :
777
Abstract :
In this paper, a new algorithm based on the active contour model for medical segmentation is presented. The deformable contour in the object is optimized by a different method of minimizing the energy function based on the greedy algorithm, a novel initialization algorithm to locate the initial contour has been developed. A new term for the internal energy and another one for external energy which solve common problems associated with classical active contour model in this type of images have been introduced. The internal energy of the model depends on local behavior of the contour, while the external energy is derived from image wavelet transform modulus maxima features. The algorithm is able to detect convex and concave object, Examples are presented to demonstrate the efficiency of the approach on spinal CT/MRI images and the results are encouraging.
Keywords :
biomedical MRI; computerised tomography; image segmentation; medical image processing; wavelet transforms; active contour model; deformable contour; external energy; greedy algorithm; image wavelet transform modulus maxima feature; initialization algorithm; internal energy; medical image segmentation; spinal CT-MRI imaging; Active contours; Biomedical imaging; Computed tomography; Deformable models; Greedy algorithms; Image segmentation; Magnetic resonance imaging; Minimization methods; Optimization methods; Wavelet transforms; Active contour models; Image segmentation; Modulus maxima; Snake; Wavelet transform;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Complex Medical Engineering, 2007. CME 2007. IEEE/ICME International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-1077-4
Electronic_ISBN :
978-1-4244-1078-1
Type :
conf
DOI :
10.1109/ICCME.2007.4381844
Filename :
4381844
Link To Document :
بازگشت