Application of Artificial Intelligence in Lung Cancer Detection: The Integration of Computational Power and Clinical Decision-Making

Abstract

AbstractLung cancer is the most common cause of cancer death around the world with over 1.80 million deaths per year. About 60% of cases have been diagnosed at an advanced stage, and the 5-year survival rate falls between 10% and 20%. Since lung cancer survival is largely determined by the stage of the disease at diagnosis, developing a reliable and effective screening method for early diagnosis has been a long-term goal of lung cancer care. Through effective screening, lung cancer can be diagnosed and treated earlier, resulting in reduced morbidity and mortality. Lung cancer screening would be more accurate and efficient by incorporating an automated intelligent system that matches or surpasses the diagnostic capabilities of human experts. The use of Artificial Intelligence (AI) for lung cancer screening can yield significant benefits. Application of AI in various healthcare settings is currently proving to be a success due to its ability to emulate human cognition in different areas such as analysis, interpretation, and comprehension of complex data sets using complex algorithms and software. By applying AI to imaging diagnostics, radiologists will be less burdened, and screening for lung cancer will be more sensitive, which, in turn, will reduce the morbidity and mortality associated with lung cancer. There has been a significant amount of research using AI to develop tools for detecting and classifying lung cancer using patient data with the goal of improving outcomes. In this article, we provide an overview of the application of AI in lung cancer for improving nodule detection and classification, as well as the challenges that remain to be overcome.A comprehensive search was performed in PubMed, Google Scholar, Web of Science, IEEEXplore, and the Cochrane Database for studies published between 2011 and 2021. The search terms were “lung cancer,” “Artificial Intelligence,” “machine learning,” “deep learning,” “screening,” “detection,” and “classification.” After excluding animal studies, relevant articles reporting AI models for the analysis of chest CT scan images in the case of lung cancer detection were discussed.Based on the results of this study, there is no doubt that AI-implemented neural networks will be the future of lung cancer screening and diagnosis. In CT images, these neural networks offer great promise in detecting lung nodules with a diameter of just 3 mm. Thus, neural networks can detect lung nodules in their earliest stages, thereby improving the survival rate of the patients.In this chapter, we discuss the role of AI, especially neural networks, in lung cancer as it relates to improving screening, diagnosis, and patient’s outcomes. There is a need for future research that will investigate deep neural networks in a clinical and biological context, as well as validate these networks in prospective data.KeywordsArtificial Intelligence (AI)Computer-aided detection (CAD)Convolutional neural networks (CNNs)DiagnosisLung cancerLung cancer detectionMachine learning (ML)Pulmonary nodule