Breast cancer diagnosis based on curvelet transform and locality sensitive discriminant analysis with reduced feature set
In this work, a simple and novel approach for breast cancer diagnosis is proposed. We present a highly discriminative and simple descriptor to differentiate between malignant and benign mammogram abnormalities. The proposed method introduces the application
of digital curvelet transform and explores feature reduction properties of locality sensitive discriminant analysis (LSDA). The mammogram image is mapped to the curvelet space. Nevertheless, The direct use of curvelet coefficients increases the curse of dimensionality problem and therefore a reduction method is required to handle the huge number of features before performing the classification. LSDA can overcome this problem taking in consideration both the discriminant and geometrical structure of the data. Finally, to accomplish the classification task of the reduced feature vectors we use the nearest neighbor classifier. This work can be considered as a stepping stone for additional research in this direction. Extensive experiments on digital database for screening mammography and INbreast database, illustrate the effectiveness of the proposed method. In addition, empirical comparisons of the proposed method against curvelet transform in combination with traditional dimensionality reduction tools demonstrate that the suggested method does not only engender a more reduced feature set, but it also outperforms all the compared methods in terms of accuracy.
Regarding indexing issue:
We have provided the online access of all issues & papers to the all indexing agencies (as given on our journal home web site). It’s depend on indexing agencies when, how and what manner they can index or not. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information.