Automated analysis of cervix images to grade the severity of cancer

Abstract

This paper discusses a method to objectively analyze the severity of cancer in images obtained from cervix. We propose a novel method to identify the transformation zone in post Lugol's iodine images and acetic acid images that are obtained from the cervix to grade the severity of cancer. We segment the Lugol's iodine image to identify the abnormal tissues and map them to acetic acid images to accurately identify the abnormal tissues in post acetic acid images as well. This information is further used to obtain an opacity difference score that could be used for grading the cancer.