DocumentCode :
683764
Title :
A phase-based active contour model for segmentation of breast ultrasound images
Author :
Lingyun Cai ; Yuanyuan Wang
Author_Institution :
Dept. of Electron. Eng., Fudan Univ., Shanghai, China
fYear :
2013
fDate :
16-18 Dec. 2013
Firstpage :
91
Lastpage :
95
Abstract :
Due to the speckle noises, low contrast and blurry boundaries in breast ultrasound (BUS) images, extraction of the boundaries in BUS images is always a challenging task. To solve this problem, a novel phase-based active contour (PBAC) model is proposed. First, we utilize the local phase information and apply the phase asymmetry approach to form a new edge indicator, which dramatically increases the robustness to intensity inhomogeneous. Then, a novel phase-based edge indicator is incorporated into a various level set formulation with the local region-based segmentation energy. Experiments are performed on both synthetic and real BUS images. The results show that the proposed PBAC model outperforms the state-of-art methods both qualitatively and quantitatively.
Keywords :
biological tissues; biomedical ultrasonics; edge detection; image segmentation; mammography; mathematical analysis; medical image processing; speckle; BUS images; PBAC model; blurry boundaries; boundary extraction; breast ultrasound image segmentation; intensity inhomogeneity; level set formulation; local phase information; local region-based segmentation energy; low contrast boundaries; phase asymmetry approach; phase-based active contour model; phase-based edge indicator; robustness; speckle noises; Active contours; Image edge detection; Image segmentation; Level set; Nonhomogeneous media; Solid modeling; Ultrasonic imaging; active contour; breast ultrasound (BUS) images; image segmentation; level set; local phase;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical Engineering and Informatics (BMEI), 2013 6th International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4799-2760-9
Type :
conf
DOI :
10.1109/BMEI.2013.6746913
Filename :
6746913
Link To Document :
بازگشت