A new method based cancer detection in mammogram textures by finding feature weights and using Kullback-Leibler measure with kernel estimation

Abstract

In these days, there are many various diseases, whose diagnosis is very hardly. Breast cancer is one of these type diseases. In this article, an diagnostic method is based on minimum-Redundancy-Maximal-Relevance m(RMR) and Kullback-Leibler (KL) Classifier for diagnosis of breast cancer. This diagnosis method is called as m(RMR)_KL. Minimum-Redundancy-Maximal-Relevance m(RMR) is used for feature selection. In this study the aim is to determine possibility of suspicious masses in mammogram. With this aim, probabilistic values of suspicious masses in the image are found via exponential curve fitting and texture features in order to find weight values in the objective function. Results are indicated on a scale to eliminate the suspicious lesions. Afterwards, images are classified as normal, malign, and benign by utilizing Kullback Leibler method. Here, 3 x 126 mammography images set selected from Digital Database for Screening Mammography (DDSM) are used, and severity of disease is probabilistically estimated. ROC analysis has been carried out to estimate the performance of the approach. Efficiency of the improved m(RMR)_KL method was tested as 98.3% accuracy diagnosis was obtained and it is very promising compared to the previously reported classification techniques. Thus, it is considered that workload of clinicians shall be reduced by easily eliminating suspicious images out of many mammography images. (C) 2015 Elsevier GmbH. All rights reserved.