Synergistic clinical trials with CAD systems for the early detection of lung cancer

Abstract

Background Nowadays, lung cancer catches the attention of medical and social communities in the recent years because of its high frequency allied with difficult treatment. Developing an effective computer-aided diagnosis (CAD) system for lung cancer is of great clinical importance and can increase patient’s chance of survival. A systematic screening of lung cancer using computed tomography (CT) images is reliable enough to find lung tumors in their early stages. Materials and Methods We conceded the available lung cancer images and its database to preprocess so as to achieve more quality and accuracy in our experimental results. After identifying the size and grade of the lung tumors along with the attributes collected from the patient’s history, we can endure classification. Significant frequent patterns are revealed using Vote, SMO, IBk, multilayer perceptron (MLP), J48, ZeroR, and Naïve Bayes classifier. Results Finally, J48 classifier outperforms, which yields 99% of correctly classified instances. Our main objective is to predict the lung cancer image to be fit in benign or malignant class.