Utilizing Convolutional Neural Networks for the Identification of Lung Cancer

Lung cancer is a disease that spreads worldwide and claims many lives each year. Applying appropriate treatments that increase the chances of patient survival requires early detection. Lung cancer identification has greatly benefited from the use of computer-aided diagnosis methods, especially when convolutional neural networks are used to analyze CT images. A methodology for creating a convolutional neural network that addresses the selection of hyperparameters is presented in this paper. A convolutional neural network is developed as a proof of concept using a small dataset using this methodology. The results support the application of the proposed methodology, resulting in a high-accuracy network that accurately classifies 97.5% of the test data that was not visible.