Lung Cancer Detection using Transfer Learning and EfficientNet B2 Architecture

The world's most common cancer-related death is lung cancer, and better patient outcomes depend on early identification. The preferred imaging technique for screening for lung cancer is computed tomography (CT); however, due to its limited sensitivity, it can potentially miss diagnosis and produce false-negative results. Some current methods employ various filter types (Gabour, Median, etc.), which might not offer the necessary precision. So, to boost accuracy and processing speed, a transfer learning (TL) method is utilized with the help of EfficientNet B2. The deep learning-based strategy for tumor identification on CT scan pictures is proposed in this paper. The suggested technique automatically extracts features from CT images and categorizes them as normal or abnormal using Transfer learning a convolutional neural network (CNN).