Abstract
Human papillomavirus (HPV) is one of the most critical viruses that integrates with human cells, and may cause various types of cancer, including cervical and oropharyngeal cancer. It encompasses a wide range of genotypes that vary in their severity and pathogenicity, which are classi@ied as low-risk that often associated with benign lesions such as warts. The high-risk types, which are strongly linked to the development of several cancers, most notably cervical cancer and the head and neck cancer. Epidemiological studies indicate that HPV infection is widespread; yet, most infections are transient and cleared by the immune system. However, in a limited percentage of cases, the infection persists, thereby increasing the risk of malignancy. Early detection and accurate diagnosis are crucial in reducing the burden of disease. Therefore, multiple techniques are employed, including molecular tests such as PCR, immunological markers such as p16, and traditional histological examination using hematoxylin and eosin (Hematoxylin and Eosin) staining. Arti@icial intelligence (AI) techniques, have emerged as promising tools by utilizing DNA data or analyzing cell images from Pap smears and histopathological images; however, challenges remain regarding the generalizability, interpretation, and integration of these models into clinical practice. This research provides a comprehensive summary of all the studies and methodologies employed in this @ield that utilize deep learning and machine learning techniques. HPV represents a global health challenge that requires integrated efforts encompassing prevention, such as vaccination, early detection, and treatment, along with leveraging modern technologies to enhance detection accuracy and improve patient outcomes.
Keywords: Cervical Cancer, High-risk HPV, Histopathology, Human Papillomavirus (HPV), Oropharyngeal Cancer.