Lesion detection and characterization in digital mammography by Bezier histograms

Author

Qi, Hairong ; Snyder, Wesley E.

Author_Institution

Dept. of Electr. & Comput. Eng., North Carolina State Univ., Raleigh, NC, USA

Volume

2

fYear

1998

fDate

29 Oct-1 Nov 1998

Firstpage

1021

Abstract

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

Keywords

image classification; image segmentation; image texture; mammography; medical image processing; principal component analysis; splines (mathematics); tumours; Bezier histograms; Bezier splines; Gaussian histograms; computer-aided mammogram diagnosis; digital mammography; high sensitivity; histogram matching; lesion characterization; lesion detection; natural threshold; principal components; region growing; segmentation; shape information; smoothed histogram; Computer errors; Histograms; Human factors; Image segmentation; Lesions; Mammography; Markov random fields; Shape; Stochastic processes; Wavelet analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE

Conference_Location

Hong Kong

ISSN

1094-687X

Print_ISBN

0-7803-5164-9

Type

conf

DOI

10.1109/IEMBS.1998.745623

Filename

745623