Integration of shape texture Feature descriptors for pulmonary Nodule classification system in lung Computed tomography images

Abstract

In medical image processing, the detection of cancer gives many benefits for humans. Cancer is a disease which spreads via the body cells and affects the human body. During our lifetime, new cells are created in our body which is controlled by the certain type of genes. If there is any change of these genes, it causes disease or cancer. Among various types of cancer, lung cancer is a leading disease / cancer in the world. To find the lung cancer, nodules are used which is placed in the lungs. The lung cancer detection is performed based on these nodules size. Computer-Aided Diagnosis (CAD) system is used for nodule detection which helps radiologists and it gives better lung cancer diagnosis. The CAD system depends on the speed and accuracy which find nodules. The output of a CAD system is taken as an option for making a decision by radiologists. For pulmonary nodule detection in lung Computed Tomography (CT) images, feature extraction offers a vital role for artificial vision implementations. The effective feature extraction is still an open and important issue in lung nodule detection and classification system. So, this research work mainly focuses on feature extraction for extracting the features and classification which satisfies the user needs. newline