Detection of Lung Cancer by Deep Learning and Machine Learning Techniques

Early detection of lung cancer cells can help in a sharp decrease in the lung cancer mortality rate hence it is an aggressive disease which carrying a dismal prognosis with a 5-year survival rate at 18%. Several computer-aided diagnosis systems have been developed to help reduce lung cancer mortality rates. Thus structural co-occurrence matrix (SCM)-based approach is used to extract the feature and to classify nodules into malignant or benign nodules and also into their malignancy level. The computed tomography (CT) scan from the lung image database consortium and image database resource initiative datasets provide knowledge concerning nodule positions and their malignancy levels is been deployed here as a model. Support vector machine is been used as a classifier which is (i) to classify the nodule images into malignant or benign nodules and (ii) to classify the lung nodules into malignancy levels (1 to 5). These experimental results reveal that the SCM successfully extracted features of the nodules from the images and, therefore may be considered as a promising tool to support medical specialist to make a more precise diagnosis concerning the malignancy of lung nodules.