Improves Treatment Programs of Lung Cancer Using Data Mining Techniques

Abstract

Lung cancer is a deadly disease, but there is a big chance for the patient to be cured if he or she is correctly diagnosed in early stage of his or her case. At a first glance, lung X-ray chest films being considered as the most reliable method in early detection of lung cancers, the serious mistake in some diagnosing cases giving bad results and causing the death, the computer aided diagnosis systems are necessary to support the medical staff to achieve high capability and effectiveness. Clinicians could predict patient’s behavior future and improve treatment programs by using data mining techniques and they can be better managing the health of patients today, in addition they do not become the problems of tomorrow. The lung cancer biological database which contains the medical images (chest X-ray) classifies the digital X-ray chest films into three categories: normal, benign and malignant. The normal ones are those characterizing a healthy patient (non nodules);, lung nodules can be either benign (non-cancerous) or malignant (cancer). Two steps are major in computer-aided diagnosis systems: pattern recognition approach, which is a combination of a feature extraction process and a classification process using neural network classifier.