Breast tumor segmentation is an essential step in a computer aided diagnosis (CAD) system. Due to the speckle noise, various size and locations of lesions, it is a challenging task to segment the tumors on breast ultrasound (BUS) image accurately. In this paper, we propose a region growing method which applies the similarity score and homogeneity based on the neutrosophic set (NS) domain to segment and detect tumors. At first, the original BUS image is transferred to the NS domain and calculate the different NS elements to obtain the NS image. Then, a thresholding method and morphology method are used to locate seed regions, and after that each seed region grows separately. The direction of region growing depends on the homogeneity vector, similarity set score vector and distance vector between the candidate points and seed regions. The suspensive condition also lies on the homogeneity vector and similarity set score vector between the seed region and the NS image. Finally, a pretrained deep learning network and transfer learning scheme are used for false positive reduction. Experiment on various clinic BUS images has suggested that the proposed method is able to segment the BUS image and extract tumor accurately.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.