Minimized Ensemble Classifiers (MEC) for the Diagnosis of Uterine Cervical Cancer using Papanicolau Smear Image Database

Uterine Cervical Cancer is one of the leading causes of increasing mortality rate among women worldwide. The cancer could be completely cured with appropriate diagnosis at an early stage using the Papanicolau (Pap) smear image test. A system which is capable of classifying with high level of accuracy is in demand. To address this challenge, this paper uses a minimized ensemble classifier in which five classification algorithms are used. The classification robustness of these classifiers are fused together as a minimized ensemble classifier system and the performance is evaluated using standard metrics. The outcomes achieved are compared with the existing system and it is found to be promising in classifying the 2-class problem with an accuracy of 99.12% and an overall error rate of 0.88%.