Early Detection of Cancer in MRI Lung Images using Classification and Regression Tree (CART) Method

Image mining is one the leading research area in the field of computer science. In this process, lung cancer is one of the most deadly disease in human body. It is the second most dangerous disease in the world. This work is mainly planned to increase the classification accuracy rate of lung cancer tissue classification in magnetic resonance (MR) imaging and reduce the processing time using Classification and Regression Tree (CART) method. In recent years, the detection cancer in early stage is a challenging task in the field of medical. This early identification of lung tumor can develop the chance of survival among the people. In this paper, we improved Classification and Regression Tree to identify and classify MR images and this is also indented to reduce the processing time, higher classification rate and minimum error rate. This proposed work consists of two phases, such as feature extraction, classification. In the first phase, we have obtained the features of magnetic resonance images have been reduced using Principle Component Analysis (PCA) method. In the second phase, CART method has been implemented to classify subjects as benign or malignant magnetic resonance images.