 |
    |
A Multilayered Ensemble Architecture for the Classification of Masses in Digital Mammograms
Peter Mc Leod and Brijesh Verma
Central Queensland University, Bruce Highway, Rockhampton, QLD 4702, Australiamcleod.ptr@gmail.com
b.verma@cqu.edu.au
Abstract. This paper proposes a technique for the creation of a neural ensemble that introduces diversity through incorporating ten-fold cross validation together with varying the number of neurons in the hidden layer during network training. This technique is utilized to improve the classification accuracy of masses in digital mammograms. The proposed technique has been tested on a widely available benchmark database.
Keywords: Ensemble classifiers, committee of experts, neural networks, digital mammograms
LNAI 7691, p. 85 ff.
lncs@springer.com
© Springer-Verlag Berlin Heidelberg 2012