Detection and Identification of Cervical Cancer on Elephant Herding Optimization on Convolutional Neural Network

Abstract

Cervical cancer is the leading cause of cancer death in women. Early diagnosis and treatment of this tumor can reduce its complications. The purpose of this article is to diagnose and classify cervical cancer using an Elephant Herd Optimization on the basis of Convolutional Neural Network (EHCNN). First, Image database of open-source system is collected. The images are sent to a central filter to remove unwanted noise. This proposed study helps in prediction and classification of cervical cancer at early stages to avoid complications. The EHCNN algorithm's weight for classifying normal, abnormal, and the type of aberrant cells was enhanced as part of this classification. This proposed classification method is used to improve the application of CNN design parameters and is selected with the help of EHO. The project shows that the images can be identified using deep learning techniques and they may have some relevance to diagnose early cervical cancer. The impact of the proposed technical results is evaluated and used for performance and image quality analysis in terms of accuracy metrics.