Wavelet transform matched filters for the detection and classification of microcalcifications in mammography

Author

Strickland, Robin N. ; Hahn, Hee Ll

Author_Institution

Dept. of Electr. & Comput. Eng., Arizona Univ., Tucson, AZ, USA

Volume

1

fYear

1995

fDate

23-26 Oct 1995

Firstpage

422

Abstract

We show that a biorthogonal spline wavelet closely approximates the pre-whitening matched filter for detecting microcalcifications in digitized mammograms. The wavelet transform acts like a bank of such detectors operating at multiple scales. Object shape computed from zero-crossings produced by the same wavelet helps to reduce false alarms. Detection performance is tested on 40 mammograms digitized at 100 μm resolution

Keywords

diagnostic radiography; image classification; image reconstruction; image segmentation; image texture; matched filters; medical image processing; splines (mathematics); wavelet transforms; biorthogonal spline wavelet; breast cancer; digitized mammograms; false alarms; mammography; microcalcification classification; microcalcification detection; multiple scales; object shape; pre-whitening matched filter; resolution; wavelet transform matched filters; zero-crossings; Breast; Frequency; Mammography; Markov processes; Matched filters; Noise shaping; Shape; Spline; Testing; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 1995. Proceedings., International Conference on

Conference_Location

Washington, DC

Print_ISBN

0-8186-7310-9

Type

conf

DOI

10.1109/ICIP.1995.529736

Filename

529736