Lung Nodule Detection Using Hybrid Patch Intensity Based Fruit Fly Optimization Segmentation And Classification

Abstract

One of the most prominent reasons of deaths due to cancer is Lung Cancer, as every year it demands lives of millions of people. The influence of carcinoma in lungs varies depending on the stages of its existence. To decrease the mortality rate of the cancer patients, cancer must be detected at the early stages. The pulmonary nodule presence is the possible indication of carcinoma. There is a huge demand for the techniques which detect the cancer at early stages. A lot of methods are available in the literature for diagnosing lung cancer and also for the classifications with some overheads. In the recent years, there is a decent advancement in the imaging technique and humungous advancement in the computer science stream, which will help to diagnose cancer in lung images at the earlier stages with this, the mortality rate of the cancer patients can be reduced and also the treatment for the cancer patients will be decided. One of the most widely used techniques for the detection of the lung cancer is Computer Aided Diagnosis System. newlineIn many cases, radiologists require help from the technical side for confirming the diagnosis made by them due to contradictions and complications while interpreting the data. Diagnosis of the nodules in the lung CT images has become one of the most prominent tasks which can be done with the help of important tool called image processing technique, for the last few years more research was carried out on developing a CAD system which interprets the lung CT images and can detect nodules in it. Most of the CAD systems in the literature are suffering with False Positive Rates and low sensitivity and also most of the CAD systems in literature are limited to the diagnosis of Non-Small Cell Lung Cancers. newlineIn this work, the foremost focus is to help the physicians by developing a CAD technique that diagnoses lung carcinoma in the CT lung images. This fully automated system can detect the presence of nodules on the CT lung images and also reduces false positives, which helps the radiologist