Improved Bald Eagle Search Optimization With Deep Learning-Based Cervical Cancer Detection and Classification

Abstract

Cervical cancer (CC) is the fourth most popular cancer affecting women worldwide. Mortality and incidence rates can be consistently enhancing, particularly in emerging countries, because of the lack of screening services, lack of awareness, and restricted qualified experts. CC has screened utilizing human papillomavirus (HPV) test, Papanicolaou (Pap) test, histopathology test, and visual inspection after application of acetic acid (VIA). Intra-and Inter-observer variability can take place in the manual analysis method, resulting in misdiagnosis. Previous studies have exploited either deep learning (DL) or machine learning (ML) approaches, the preceding one could not be efficient as it needs segmentation and attaining hand-crafted features that utilize critical stage. Artificial Intelligence (AI) based computer-aided diagnoses (CAD) methods are generally explored for identifying CC for enhancing the standard testing method. This manuscript offers an Improved Bald Eagle Search Optimization with Deep Learning based Cervical Cancer Detection and Classification (IBESODL-CCDC) algorithm. The drive of the IBESODL-CCDC algorithm lies in the automated classification and detection of CC. In the presented IBESODL-CCDC technique, a contrast enhancement process takes place to enhance the image qualities. In addition, the IBESODL-CCDC technique utilizes a modified LeNet model for the feature extraction model. For CC detection, the IBESODL-CCDC technique applies an attention-based long short-term memory (ALSTM) network. A wide-ranging experiment was applied to validate the greater outcome of the IBESODL-CCDC technique. The experimental values highlight the remarkable performance of the IBESODL-CCDC algorithm with other recent systems.