Bernard, Sylvain, Muller, Serge, and Onativia, Jon
Abstract
In this paper, we present a method for microcalcification detection on filtered back-projection reconstructed slices of breasts acquired on a Digital Breast Tomosynthesis system. Performance of the algorithm has been evaluated on a ground-truth database composed of 50 cases (13 lesions, 37 normals). Since the method is derived from an algorithm developed to detect microcalcifications on 2D digital mammograms, a comparison in terms of performances is provided leading to a discussion for future improvements. [ABSTRACT FROM AUTHOR]