Effective Deep Ensemble Methods for the Detection of Lung Cancer with SMOTE

Abstract

Lung cancer is a death-causing disease that indicates the existence of pulmonary nodules in the lung. It is typically caused by increased cancer cells in the lung. Lung nodule detection has a major role in screening and detecting lung cancer images in computed tomography (CT) scans. Lung cancer deserves more attention in human disease investigation methods due to its substantial influence on both males and females, leading to around five million deaths each year. The patient’s survival ratio is increased as lung cancer detects at an early stage. This work presents a novel and unique transfer learning model based on deep ensemble techniques for efficient lung cancer detection. A chest CT scan dataset is needed to detect lung cancer. The presented technique diagnoses lung cancer efficiently with high accuracy. The primary objective of this research is to detect lung cancer at an early stage by evaluating the efficacy of deep ensemble classification models. The proposed model achieves outstanding results when compared with state-of-the-art methods to classify lung cancer. The experimental results show significant performance of the presented model.