Automated Cervical Cancer Detection using Level Set segmentation and Two-Level Cascade Classification

Pap smear test has been broadly used for detection of cervical cancer. Cervical cancer ranks as the fourth most prevalent cancer affecting women worldwide and its early detection provides the opportunity to help save life. To that end, automated diagnosis and classification of cervical cancer from pap-smear images has become a necessity as it enables accurate, reliable and timely analysis of the condition’s progress. So this paper proposes a method for automatic cervical cancer detection using cervical cell segmentation and classification. In this paper, the Herlev dataset, which consists of 7 classes, is used. A single cervical cell image is segmented into cytoplasm and nucleus using Level Set method. We first proposed 24 features, including morphologic and texture features, based on the characteristics of each cell type. We then used a two-level cascade integration system of two classifiers to classify the cervical cells into normal and abnormal cells. In the first step, C4.5 is used to classify the cells into 7 different classes and in the second step Logistic Regression classify cells into normal and abnormal cells. The experiments show that the overall results of the proposed two-level classification are better than the SVM and ANN single classifiers.

Keywords:

pap smear test , cervical cancer , Level Set method , cascade classifier