Enhancing Lung Cancer Detection and Classification Using Machine Learning and Deep Learning Techniques: A Comparative Study
Amira Bouamrane, Makhlouf Derdour
Abstract
Lung cancer is a huge and serious hazard to human health, as it is one of the deadliest cancers for both men and women, owing to its ability to metastasize to other organs. Cancer detection and therapy options remain difficult for experts in this sector. To solve these difficulties, artificial intelligence approaches such as machine learning and deep learning have produced very promising and practical results. Deep learning, in particular, provides extraordinary precision and speed, allowing for the identification of even the smallest pulmonary nodules with extensive information. Several organizations have helped by offering databases to aid scientific study. Among these, the LIDC-IDRI dataset is one of the world's largest publicly available databases. CT images for 1012 patients are included. In this paper, we compared SVM, KNN, DT, and Random Forest machine learning models, as well as VGG19, EffetientNet-V2-L, Wide-Resnet-50-2-weight, and EffetientNet-B7 models.