Analysis of Cervical Neoplasia with Artificial Intelligence

Cervical cancer is the second cancer that is the most common one and causes death in women after breast cancer. Still, it is a type of genital cancer that can be prevented by early diagnosis with screening tests. Most cervical cancers 99.7% are related to the Human Papilloma Virus (HPV), and today, it is known that HPV must definitely exist for the development of cervical cancer. Actually, HPV-16 and HPV-18 have been found positive in 70% of the patients. In this study, data on the cervical cancer-HPV relationship, dysplasia, and cancer development rates have been analyzed using retrospectively investigated artificial intelligence techniques. This study has been carried out with 1147 patients selected from a total of 2850 patients aged 20–59 years who applied to the Department of Obstetric and Gynecology of the University of Health Sciences (SBU) Ankara Training and Research Hospital in Turkey between June 2018 and June 2020. The clinical information in the file of each selected patient has been queried, all cases have been examined by the same obstetrician, and a pap test has been performed. The Pap test has been reevaluated by the same specialist pathologist using the Bethesda (2001, modified 2014) System. In addition, the results of HrHPV DNA (HPV types 16,18,31,33,35,39,45,51,52,56,58,59,66,68) tests performed on the patients included in the study have been reviewed. The results of the patients who were found to have Low-Grade SIL (LGSIL) and High-Grade SIL (HGSIL) cancer findings in the Pap test and underwent HPV DNA analysis and other data in the file information were interpreted with artificial intelligence techniques. It has been found that of 1147 patients, 147 have LGSIL, 165 have HGSIL, and 97 have cervical cancer. It has been observed that there has been the highest incidence of LGSIL, HGSIL, and cancer in the 30–39 age group, while the rate of these diseases is lower in the 50–59 age group compared to other groups. In addition, all patients with SIL and cancer are smokers. 231 of 1147 patients have positive HPV DNA tests. The highest positive HrHPV DNA has been seen in 90 patients aged 30–39 years. In 71 of 97 patients with cervical cancer, HrHPV DNA positivity has been reported. Moreover, in this study using artificial neural networks, it has been measured that the accuracy value is 95%, the precision value is 100%, the recall value is 92%, and the F1-Score value is 96%.