Classification of Cervical-Cancer Using Pap-Smear Images: A Convolutional Neural Network Approach

Cervical cancer is the second most common and the fifth deadliest cancer in women. In this paper, we propose a deep learning approach for detecting cervix cancer from pap-smear images. Rather than designing and training a convolutional neural network (CNN) from the scratch, we show that we can employ a pre-trained CNN architecture as a feature extractor and use the output features as input to train a Support Vector Machine Classifier. We demonstrate the efficacy of such a new employment on the Herlev public database for single cell papsmear, whereby the experimental results show that our proposed system neatly outperforms other state of the art methods.