Detection of Lung Cancer Using Backpropagation Neural Networks and Genetic Algorithm

Abstract: Carcinoma also known as Cancer is the commonly growing and most dangerous disease occurred in human species. Lung Carcinoma is one of them. It is a disease that occurs due to uncontrolled growth of cancerous cells in the tissues of the lungs. Prior diagnosis of the disease saves huge number of lives, failing in which may lead to other severe problems causing sudden fatal death. Motive of this system is to automate the detection process so as to perform advanced detection of the disease in its early stage. A measure for early diagnosis mainly includes X-rays, CT images etc. In this system firstly we use techniques such as Data Preprocessing, Training and testing of samples that are necessary for the task of medical image mining. A powerful learning model i.e (BPNN) is used for classification which would classify the digital X-ray, CT-images, MRI�s, etc as cancerous or non-cancerous. Further Genetic Algorithm will be used that would extract feature on the basis of the fitness function. The selected feature will help to detect the stage of cancer by measuring the size of the feature and using this measurement the stage of cancer will be decided. This system will support to make an relevant decision about a patient�s state.

Keywords: Backpropagation, Neural Networks, Classification, Genetic Algorithm, Medical Image Mining.