A Lung Cancer Detection System Based on Convolutional Neural Networks and Natural Language Processing

Weixin Wang, Qianli Ma, Jiahao Chen

Abstract: Lung Cancer has long been regarded as one of the most threatened diseases to human beings, and detecting early malignant tumors is of vital importance for treatment. Contemporarily, Radiology departments in hospitals usually have to deal with multiple CT images to carry out the detection, which is a huge workload for doctors. Here, we propose a novel system to help with lung cancer detection. Specifically, deep feature based convolutional neural networks (CNN) is applied to classify lung cancer tumors, realizing an accuracy of 88%. Moreover, a chatbot based on natural language processing (NLP) technology is embedded into the system to provide immediate knowledge and information. These results shed light on how doctors’ workload might be reduced to a considerable extent.