A New Method for Lung Nodule Detection Using Deep Neural Networks for CT Images

Lung cancer has been the leading cause of cancer-related deaths worldwide. Early detection of cancer greatly increases the chances for successful treatment and diagnosis and reduce morbidity. In this paper, we propose a lung cancer detection and prediction method using the deep neural network. The proposed technique is used to detect the lung cancer in its early stage and also to predict the lung cancer. The proposed method works in two phases: Firstly, the CT images are preprocessed through blurring and thresholding to improve the quality and simply the images. Then split the left and right lung for reducing the pattern as well as reducing the time complexity and increasing the accuracy that is used in deep neural network. In order to verify our proposed method we use MobileNet, VGG-8 and Inception-v3 deep neural network models as an image classifier. From the experimental analysis, we illustrate that the MobileNet model gives the best performance while splitting the CT images than the VGG-8 and Inception-v3.