Cervical Cancer Automated Screening Module - CervCancerScreening

Abstract

There are more than 100 types of cancer and they are the leading causes of deaths all over the world. Many of them are fatal but many other can be treated if they are diagnosed early on, like the Cervical Cancer. This being said, the Cervical Cancer is still the second leading cause of female deaths in developing countries. The reason behind this is because of the lack of early diagnosis and sometimes even because lack of information. Since the women in this countries do not have gynecological surveillance, mainly because its cost, chances are high that the cancer grows to a state that can no longer be treated. When the surveillance actually exists, it is very expensive since the samples of the tests made need to be sent to another country with access to specific equipment. With the increasing techonological development, it is possible to create a less expensive tool to try to supress some of this problems. The tool will be used to assist the medical staff so that they have a first overview of the patient's cervical cancer results. Nevertheless, the ultimate decision on whether a person actually has or does not have cancer comes from the doctor.The objective of this thesis is to develop a tool that can support and be used by developing countries health facilities, so that they have the first impression about the situation of the patient by analysing the pictures of the samples gathered during the appointment. The main focus of the tool is to analyse and identify specific characteristics of the sample pictures in order to be able to give feedback about the existence of the cancer. In the end, the main objective is to spare those facilities sending the samples to other places and spend more money than they should, as the tool would give some insights about the risks and the need of further analyses.