Cervical cancer is a common malignancy encountered among many people in recent days. Early identification and detection can significantly aid in the management and treatment that follows. As advances in artificial intelligence (AI) and deep learning (DL) approaches have been made, an expanding range of deep learning-based (computer-aided detection (CAD) techniques have been used for cervical cytology screening. Firstly, we present a brief overview of the biological and medical information related to cervical cytology, as a thorough understanding of biomedicine can significantly impact the advancement of computer-aided detection (CAD) systems. Next, an extensive analysis of few-shot learning analyzed by many researchers is discussed. Furthermore, a synopsis of image analysis methods and uses is offered, such as cervical entire slide image diagnostic, cell region segmentation, detection of aberrant cells or areas, and identification of the cervical cell. In conclusion, they address current challenges and prospective avenues for further investigation into automated cervical cytology screening research.