Segmentation and analysis of breast tumors on ultrasonography

Abstract

Ultrasonography provides some criteria which help the physicians decide whether a certain solid tumor is benign or malignant. However, it is one of the most difficult types of images to assess. The automatic segmentation and analysis of ultrasonography can help the physicians by providing some techniques and measures to classify the tumors. In this paper, we present a new approach in the segmentation of ultrasound images of breast nodules, based on active contours technique. Moreover, we present a common framework for the extraction of a set of robust, reproducible and precise parameters, by means of computer vision tehniques, such as ellipse location, corner extraction or gradient estimation.