Early diagnosis is crucial for improving lung cancer prognosis, a leading cause of cancer-related deaths. Lung cancer includes small cell lung cancer (SCLC, ~15% of cases) and non-small cell lung cancer (NSCLC, ~80–85%). Prognosis depends on the stage at diagnosis; the 5-year survival rate is 65% for localized NSCLC but only 9% for distant-stage disease. Radiologists face challenges distinguishing benign from malignant pulmonary nodules on CT scans.This review explores deep learning (DL) methods, including multi-view Convolutional Neural Networks (CNNs) and 3D models for nodule segmentation, emphasizing volumetric assessments for malignancy prediction. CNNs effectively analyze CT data, achieving 94.2% sensitivity with 1.0 false positives per scan in lung nodule detection. DL enhances diagnostic accuracy, reduces radiologist workload, and enables earlier lung cancer detection. Further research is needed to improve model adaptability across diverse clinical settings.