Cervical cancer, a cancer prevalent among women worldwide, can be managed with early detection and diagnosis. This paper, leveraging the advancements in artificial intelligence (AI) techniques, presents a systematic literature review (SLR) for the application of Deep Learning (DL) and Machine Learning (ML) models for segmenting and classifying single cells and whole-slide imaging (WSI) in Papanicolaou (Pap) tests. The study underscores the use of Convolutional Neural Networks (CNN) for feature extraction and classification, as well as other ML models such as support vector machines (SVM) and Random Forests for classification. However, the focus of most studies on segmenting and classifying individual cervical cells neglects the crucial aspect of whole-slide image (WSI) classification, primarily because of the scarcity of accessible datasets. The study concludes that the number of studies proposing methodological approaches for automated Pap test reading has significantly increased since 2019, underscoring the need for robust solutions to enhance the sensitivity and efficiency of the Pap test, thereby leading to improved rates of early detection of cervical cancer.