Title :
A wavelet neural network for the detection of microcalcifications in multiscale enhanced digitised mammograms
Author :
Aydin Alayhoglu, B. ; Aghdasi, Farzin
Author_Institution :
Dept. of Electr. Eng., Univ. of the Witwatersrand, Johannesburg, South Africa
Abstract :
The detection of microcalcification clusters (MCCs) during mammographic examination is vital for the correct diagnosis of breast cancer in asymptomatic women. Automated detection methods augment the diagnostic process by providing the radiologist with a second opinion. Detection schemes making use of a neural network classifier, with input feature vectors containing spatial, spectral and multiscale image attributes, are investigated. A wavelet-based image enhancement technique is implemented to improve the detection performance
Keywords :
cancer; image classification; image enhancement; mammography; medical image processing; neural nets; wavelet transforms; asymptomatic women; automated detection methods; breast cancer diagnosis; detection performance; input feature vectors; microcalcification clusters; microcalcification detection; multiscale enhanced digitised mammograms; multiscale image attributes; neural network classifier,; spatial image attributes; spectral image attributes; wavelet neural network; wavelet-based image enhancement technique; Attenuation; Breast cancer; Cancer detection; Image enhancement; Intelligent networks; Mammography; Neural networks; Radiography; Tumors; Wavelet transforms;
Conference_Titel :
Africon, 1999 IEEE
Conference_Location :
Cape Town
Print_ISBN :
0-7803-5546-6
DOI :
10.1109/AFRCON.1999.820874
