LUNG CANCER DETECTION USING DEEP LEARNING

Abstract

In this paper we focuses on detecting the lung cancer at the very early stage using advanced deep learning architecture and transfer learning methods. Lung cancer is one of the leading causes of death in both men and women around the world, with an estimated five million deaths each year. In the detection of lung disorders, a computed tomography (CT) scan may be very helpful. The key goal of this project is to identify cancerous lung nodules in an input lung picture and to distinguish lung cancer and its magnitude. This study employs innovative Deep learning approaches to detect the presence of cancerous lung nodules. The best feature extraction techniques are used in this study, including the Histogram of Oriented Gradients (HoG), wavelet transform-based features, Local Binary Pattern (LBP), Scale Invariant Feature Transform (SIFT), and Zernike Moment. The VGG16 algorithm is used to find the best function after extracting texture, geometric, volumetric, and strength features. Finally, Deep learning is used to classify these functions. The computational complexity of CNN is reduced thanks to a new VGG16.