A Customizable Intelligent System for Cervical Cytology Image Classifications

Cervical cancer is an important health concern and early detection is crucial for effective treatment. In this study, we propose a novel approach to computer-assisted cervical scans that uses a custom-based deep learning architecture, called Baby-Fortune Intelligent System (BFIS), designed for multiple binary classifications of the image of cervical cells. BFIS is trained on comprehensive publicly available data sets that include Pap smear captures. Comparative analysis with existing screening tools highlights the superior performance of the BFIS in terms of sensitivity, precision, and overall predictive accuracy. Our results underscore the potential of the proposed model as a valuable tool for cervical screening, which offers enhanced results and efficiency in identifying people at increased risk of cervical malignancies.