Image Processing Reference
In-Depth Information
Automatic mass segmentation
method in mammograms
based on improved VFC
snake model
Xiangyu Lu; Min Dong; Yide Ma; Keju Wang School of Information Science and Engineering, Lanzhou University, Lanzhou, China
Mammography analysis is an efficient way for the early detection of breast cancer. In this chapter, we
present an integrated method for mass auto-segmentation in breast. First of all, the local threshold meth-
od, rough set theory, and morphological filter are used to remove the label and enhance the mammo-
gram. Second, we apply the Hough transformation algorithm on the preprocessed image and locate the
lesion as an approximate parametric circle which would be used as the initial contour of Snake mod-
el followed by. Finally, the mass boundary is accurately segmented based on coarse localization. This
approach is tested on digital database for screening mammography and mammography image analysis
society database and the performance is evaluated from three aspects: detection rate, area-based accur-
acy, and distance-based boundary similarity measures based on manual-segmented results. By compar-
ison, we find that our improved method has higher detection rate and the segmented contours are much
closer to the actual area of objects. The promising results indicate that our approach can provide some
theoretical basis for computer-aided image detection system.
Early breast cancer detection
Mass segmentation
VFC Snake model
Authors would like to thank the retrieval of DDSM and MIAS database from the Internet for
the experiments of this chapter. This work is jointly supported by the National Natural Science
Search WWH ::

Custom Search