A Review on Lung Cancer Segmentation and Classification Using Deep Learning Techniques

Lung cancer is a leading cause of mortality, making early detection critical for better outcomes. This review explores advancements in deep learning (DL) for lung cancer segmentation and classification, focusing on recent techniques and their impact. DL methods, particularly convolutional neural networks (CNNs), have enhanced diagnostic precision by automating the detection process. Segmentation models like U-Nets and 3D-CNNs have achieved strong performance in delineating lung nodules, while hybrid models and attention mechanisms have further improved results. Despite progress, challenges remain in computational demands and generalizability across clinical settings. Classification models using CNNs and transfer learning have shown very high accuracy, in distinguishing malignancies. However, their resource intensity poses limitations for widespread use.Future direction should focus on making DL models more efficient, adaptable, and supported by diverse datasets to facilitate clinical integration and improved patient care.