International Journal of Computer Theory and Engineering

Abstract—This paper proposes a method that can enhance the performance of Computer Aided Diagnosis (CAD) by automatically detecting and classifying the microcalcifications (MCs) in mammogram image accurately and efficiently using multi statistical filters and wavelet decomposition transform. The proposed method is divided to two main stages. In first stage, the potential MCs region (PMR) is detected based on visual characteristics of the MCs in the mammogram images. Then wavelet decomposition transform is implemented to classify the PMR to true positive and false positive regions based on extraction four wavelet features for the mammogram image. This novel method was found to be sensitive in detecting MCs in mammogram images by achieving a high true positive percentage of 98.1% and a low false positive rate 0.63 cluster/image for both MIAS and USF databases.

Index Terms—Mammogram, microcalcifications, wavelet transform, wavelet features.

Ayman A. AbuBaker is with the Electrical and Computer Engineering Department, Applied Science University, Amman, Jordan (e-mail: a_abubaker@asu.edu.jo).

[PDF]

Cite:Ayman A. AbuBaker, "Automatic Microcalcification Detection Using Wavelet Transform," International Journal of Computer Theory and Engineering vol. 7, no. 1, pp. 40-45, 2015.