THE APPLICATION OF DEEP NEURAL NETWORK FOR BREAST CANCER CLASSIFICATION

Abstract

Breast cancer is one of the most common cancers, especially for women. Early detection of breast cancer may increase the survival rate of patients significantly. Detecting breast cancer from breast image can be done by classification process. There are so many classification models which can be used for the classification process, such as neural network, fuzzy, neuro fuzzy, wavelet neural network, wavelet neuro fuzzy, etc. This research propose one of the neural network variant, i.e. Deep Neural Network. This kind of neural network model is using at least two hidden layers on the network. The more hidden layers used the deeper the neural network will be. The architecture of Deep Neural Network used in this research is feedforward network. Classification of breast tumor using Deep Neural Network model provides results with sensitivity, specificity, and accuracy were respectively 100%, 100%, and 66.67% for training data and 100%, 20%, and 60% for testing data.

Keywords: breast cancer, Deep Neural Network