Using Kohonen Artificial Neural Network to Cluster and Visualize Risk Factors for Lung Cancer

Abstract

Lung cancer is a complex disease with multiple risk factors, including smoking, exposure to second-hand smoke, alcohol consumption, air pollution, occupational exposures, genetic predisposition, age, and previous lung disease. Artificial Intelligence (AI) has emerged as a promising tool for clustering risk factors related to lung cancer, improving diagnostic efficiency, providing optimal treatment, and assessing prognosis. However, only a few studies have used clustering algorithms to identify modifiable risk factors for lung cancer. In this study, we have implemented the Kohonen Artificial Neural Network-also known as the Self-Organizing Map (SOM)-to cluster and analyze a lung cancer database of 1,000 records and six attributes (risk factors). The SOM methodology is a powerful tool for clustering and scaling up data, and it has proven to be a versatile and effective technique for solving complex problems and simplifying data. Our study demonstrates the potential of SOM to revolutionize how we understand and manage lung cancer risk factors, which is necessary to improve the prevention, early detection, and treatment of this fatal disease.