Computer aided diagnosis system based on machine learning techniques for lung cancer

Cancer is a leading cause of death worldwide. Lung cancer is a type of cancer that is considered as one of the most leading causes of death globally. In Malaysia, it is the 3rd common cancer type and the 2nd type of cancer among men. In this paper, machine learning techniques have been utilized to develop a computer-aided diagnosis system for lung cancer. The system consists of feature extraction phase, feature selection phase and classification phase. For feature extraction/selection, different wavelets functions have been applied in order to find the one that produced the highest accuracy. Clustering-K-nearest-neighbor algorithm has been developed/utilized for classification. Japanese Society of Radiological Technology's standard dataset of lung cancer has been used to test the system. The data set has 154 nodule regions (abnormal) and 92 non-nodule regions (normal). Accuracy levels of over 96% for classification have been achieved which demonstrate the merits of the proposed approach.