Variable hidden neuron ensemble for mass classification in digital mammograms

This paper proposes a new ensemble technique for the classification of masses in digital mammograms based on neural networks with variable hidden neurons which are combined with hierarchical fusion. The main focus is introducing diversity into an ensemble network by varying the number of neurons in the hidden layer of the neural networks and ten-fold cross validation. The novelty of the proposed ensemble lies in the creation of diverse neural networks and combining the best performers using hierarchical fusion.