Research Challenges in Cervical Cancer Segmentation and Classification Using Colposcopy Images

Over the past three years in Indonesia, there have been 36,633 new cases of cervical cancer, the country’s second most common cause of death. Out of those cases, 21,003 people have died. The total number of deaths from cervical cancer in the past five years is 92,930. Potential deaths from cervical cancer can be prevented by early detection, such as pap smear, colposcopy, and biopsy. Experienced experts can only perform some of these measures, but the number of experts is currently 430 based on Indonesian Ministry of Health data. This study reviews several studies on cervical cancer detection using deep learning and machine learning models on colposcopy images. Since, there are still challenges to determine standard colposcopy images and overcome the interference of reflected light around the cervical lesion area. Therefore, this study will be helpful for researchers to consider practical and efficient pre-processing, feature extraction, segmentation, and classification.