Abstract
Sonography is gaining popularity as an adjunct screening technique for assessing abnormalities in the breast. This is particularly true in cases where the subject has dense breast tissue, wherein widespread techniques like Digital Mammography (DM) fail to produce reliable outcomes. This article proposes a novel and fully automatic methodology for breast lesion segmentation in B-mode Ultra-Sound (US) images by utilizing region, boundary and shape information to cope up with the inherent artifacts present in US images. The proposed approach has been evaluated using a set of sonographic images with accompanying expert-provided ground truth. © 2010 Springer-Verlag.
Author supplied keywords
Cite
CITATION STYLE
Massich, J., Meriaudeau, F., Pérez, E., Martí, R., Oliver, A., & Martí, J. (2010). Lesion segmentation in breast sonography. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6136 LNCS, pp. 39–45). https://doi.org/10.1007/978-3-642-13666-5_6
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
