Computer Aided Diagnosis Of Cervical Cancer Cells Using Deep Learning Methods

This paper shows that the deep learning method can be applied to providing computer-aided diagnosis (CAD) of cervical cancer cells. This research studies the effect of CervicalNet, a dedicated deep neural network algorithm on Detecting and categorizing cervical cancer cells through the Mendeley Liquid-Based Cytology (LBC) dataset. This disease affects women across the world, ranks as the 4th common malignancy in women; early detection improves patient care outcome. The traditional techniques are always based on manual examination of cell samples by cytopathologists, procedures that are tedious and susceptible to human errors. This work illustrates that deep learning particularly CervicalNet, can assist with automating and improving detection in the diagnosis of cell images, by accurately categorizing them as well as identifying cancerous abnormalities with high precision. The aim is to highlight the efficiency of CAD Systems in reducing diagnostic time and improving accuracy in cervical cancer screening.