Abstract

Lung cancer is one of the most occurring cancers among both men and women [1]. Detection of cancer at the early\nstage is the only method to improve the survival rate. CT images of lung are helpful in the diagnosis of lung cancer.\nDoctor analyses the CT image and predicts the presence of cancer nodule. This manual detection can also result in\nfalse detection. So in order to overcome this problem a computerized method for cancer detection is needed. Image\nprocessing technique can be used for this purpose. By using the following image processing techniques lung cancer detection\ncan be developed further. Lung cancer detection system has three steps to detect the presence of cancer nodule in lung. Preprocessing\nstage, segmentation and feature extraction stage. Pre-processing step includes image enhancement using Gaussian\nfilter and image segmentation. Enhanced CT image of lung is then fed to segmentation phase. From the segmented output,\nfeatures are extracted. By using these extracted features the lung is differentiated as normal lung or cancerous lung. And\nfurther the cancerous tumor can be differentiated as Benign and Malignant.