An Efficient Method for Lung Cancer Image Segmentation and Nodule Type Classification Using Deep Learning Algorithms

Lung cancer is one of the types of cancer in the world. It is a type of disease that develops from control and forms abnormal cells in the lungs. These cells do not function like other normal cells due to deoxyribonucleic acid (DNA) mutation by various genetic factors. However, early detection and treatment of cases can reduce the risk of cancer mortality. In turn, the use of models based on convolutive neural network (CNN) architecture in the field of medical imaging diagnostics is widespread, but these architectures have different results in terms of diagnostic accuracy. In this paper, we propose an efficient method for CT lungs images segmentation and nodule type classification using deep learning algorithms based on convolutional neural network architecture with two fundamentally distinct deep learning algorithms, U-NET for segmentation and U-Net for lung nodules classification type.