FEATURES EXTRACTION FOR MICROCALCIFICATION CLUSTERS CLASSIFICATION IN DIGITAL MAMMOGRAMS

Arianna Mencattini, Giulia Rabottino, Marcello Salmeri, Federica Caselli,Roberto Lojacono

Abstract

Breast cancer is the first leading cause of death by cancer for women. To increase the survival rate it is necessary to detect lesions as soon as possible. Most early breast cancer can be diagnosed by detecting microcalcification clusters in mammographic images. The clusters appear as groups of small, bright particles with arbitrary shapes and distribution. Because of human factors such as subjective or varying decision criteria, distraction by other image features, large number of images to be inspected, or simple oversight, some diagnosis are missed.In this paper, we propose a method to classify clusters of microcalcifications characterizing the lesion by the extraction of geometrical (2D) and textural (3D) features. Then, through a statistical analysis of these features, we can choose the most discriminating between benign and malignant lesions and so design the classifier.

Download

IMEKO-TC4-2008-003.pdf

DOI

-

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings