Lung Cancer using Detection using CNN

Abstract

Lung cancer a major threat of human life. Computer Tomography (CT) scan images are examined by the physician for identifying the cancer for patients. This examination when taken by the expertise can be able to identify the disease. When the tumor is small sometimes, the expertise may fail to notice and moreover, the manual examination is time consuming and considering the lack of experts, the artificial intelligence is widely used in the disease classification. When lung tumors is identified in earlier stages it can be diagnosed and increase the life of patients. In the artificial intelligence domain, deep learning algorithms perform accurately on disease classification from CT scan images. The proposed work is aimed at lung cancer as binary classification with normal and abnormal classes from CT scan images using deep learning algorithms. The framework developed for lung cancer identification using Convolutional Neural Network (CNN) algorithm and this application gets CT scan image from user and classifies it. Experimental results show that the accuracy achieved with the proposed CNN algorithm is 97.10%. Lung cancer is the principal cause for cancer-related death.Lung cancer can initiate in the windpipe, main airway or lungs. It is caused by unchecked growth and spread of some cells from the lungs. People with lung disease such as emphysema and previous chest problems have more chance to be diagnosed with lung cancer. Over usage of tobacco, cigarettes and beedis, are the major risk factor that leads to lung cancer in Indian men; however, among Indian women, smoking is not so common, which indicate that there are other factors which lead to lung cancer. Other risk factors include exposure to radon gas, air-pollutions and chemicals in the workplace. Lung cancer detection at early stage has become very important and also very easy with image processing and deep learning techniques. In this study lung patient Computer Tomography (CT) scan images are used to detect and classify the lung nodules and to detect the malignancy level of that nodules. In this project we are using CNN algorithm to detect Lung cancer from CT-SCAN images and to train CNN we have CT-SCAN images dataset. KEYWORD-CNN, CT, Image processing, deep learning.