Detection of calcifications in digitized mammograms using wavelet packet analysis

Author

Chiracharit, W. ; Ayudhya, P. Navaratana Na ; Chamnongthai, K.

Author_Institution

Dept. of Electron. & Telecommun. Eng., & Electr. Eng., King Mongkut´´s Univ. of Technol., Bangkok, Thailand

Volume

2

fYear

2001

fDate

6-9 May 2001

Firstpage

253

Abstract

This paper presents a method to detect calcifications, which is an early sign of breast cancer in mammograms. By wavelet packet transform, a low-frequency background is eliminated. Noises can be filtered using spatial edge images. Then the image from inverse wavelet packet transform will contain only required information. A FROC curve is used to evaluate the performance by applying it to 25 images. The result shows 93% true positive detection rate at the cost of 1.28 false positive per image

Keywords

cancer; mammography; medical image processing; wavelet transforms; FROC curve; X-ray imaging; background suppression; breast cancer; calcification detection; digitized mammogram; inverse wavelet packet transform; noise filtering; patient diagnosis; spatial edge image; Background noise; Breast cancer; Cancer detection; Frequency; Image edge detection; Low-frequency noise; Wavelet analysis; Wavelet packets; Wavelet transforms; X-ray imaging;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 2001. ISCAS 2001. The 2001 IEEE International Symposium on

Conference_Location

Sydney, NSW

Print_ISBN

0-7803-6685-9

Type

conf

DOI

10.1109/ISCAS.2001.921055

Filename

921055