Lesion detection and characterization in digital mammography by B ezier histograms

Due to some important properties of B ezier splines, they have great potential use in computer-aided mammogram diagnosis. In this paper, B ezier splines are applied in both lesion detection and characterization processes, where lesion detection is achieved by segmentation using a natural threshold computed from B ezier smoothed histogram; and lesion characterization is achieved by measuring the tness between Gaussian and B ezier histograms of data projected on principal components of the segmented lesions. Experimental results show that this approach is e cient, easy to use, and can achieve high sensitivity.