Enhancing Early Lung Cancer Detection using Deep Learning Techniques

Lung cancer is one of the most deadly and persistent diseases in the world today, making early identification of this disease crucial. This research focuses on the integration of cutting-edge deep learning technologies, namely Convolutional Neural Networks (CNNs), with modern medical insights to improve lung tumour detection using plain chest X-rays (CXR). Using a novel technique to transfer learning, we optimise our model's effectiveness for better diagnostic results in the early identification of lung cancer. Our suggested model, which combines ResNet50, EfficientNetB0, and DenseNet121, achieves outstanding accuracy (97.41%) while minimising loss (0.0803), exceeding individual models. This unique model combines the characteristics of many architectures to provide improved feature extraction and sophisticated pattern identification in medical imaging. Beyond a literature review, our study adds a real answer that has the potential to revolutionise early detection, giving a powerful weapon for combatting this lethal illness.