Computer Aided Diagnosis for Cervical Cancer Screening using Monarch Butterfly Optimization with Deep Learning Model

Cervical cancer (CC) is the commonest gynecologic malignancy around the world. Since CC is an extremely preventable disease, thus, earlier screening signifies an efficient strategy for minimizing the worldwide problem of CC. But as a result of extremely costly procedures in developing countries, scarce awareness, and access lack to medicinal centers, the vulnerable patient population could not afford to endure examination frequently. Therefore, this study presents a Computer Aided Diagnosis for Cervical Cancer Screening using Monarch Butterfly optimization with DeepLearning (CADCCSMBODL) model. The presented CADCCS -MBODL technique employs transferlearning with hyperparameter tuning strategies for CC classification. To achieve this, the presented CADCCSMBODLtechnique employs adaptive filtering (AF) with saliency based segmentation approach. For feature extraction, the presented CADCCS-MBODL technique employs EfficientNet model with MBO algorithm as a hyperparameter optimizer. Finally, extreme gradient boosting (XGBoost) classifier is applied for classification anddetection of CC. The simulation outcome of the CADCCS-MBODL technique was tested using benchmark medical database and the outcomes signified the improved outcomes of the CADCCS-MBODL system over other existing models.