Survey on Automated Colposcopy Image Classification
B. Monisha, Rama Rani, R.Kiran Manie, V.Ranichandra · Journal of Emerging Technologies and Innovative Research · 2018
Uterine cervical cancer is the second most common cancer in women worldwide, with nearly 500,000 new cases and over 270,000 deaths annually. Effective diagnosis in early stages can give women a better chance of full healing and survival. Digital imaging technologies and neural networks allow us to assist the physician with an automated Computer-Aided Diagnosis(CAD) system, In Colposcopy, epithelium that turns white after the application of acetic acid is called aceto-white epithelium. Aceto-white epithelium is one the major diagnostic features observed in detecting cancer and precancerous regions. Automatic extraction of aceto-white regions from the cervical images has been a challenging task due to specular reflection and most importantly large intra-patient variation. The proposed structure aims to ease the diagnosis of cervix cancer though an automated smart system, which allows the physician to upload the colposcopic images through a web application, track the patients Electronic Medical Records(EMR), detect the abnormality in the images and finally grade the severity level of CIN(Cervical Intra-epithelical Neoplasia) using neural networks. The proposed system aims to enhance the functioning and processing of cervical cancer screening, thus trying to make the process more reliable.