Evaluating pulmonary nodules, which may potentially indicate lung cancer, is a crucial aspect of chest imaging. The visual examination of chest radiography and chest computed tomography (CT) scans by radiologists has traditionally formed the foundation for the evaluation of lung nodules. However, this approach is hampered by high inter-reader variability, limited quantitative data, and the potential for human error. Deep learning-based artificial intelligence (AI) techniques have demonstrated exceptional performance that is comparable with, or even surpasses, that of radiologists throughout the entire process of lung nodule assessment, from detection to prognosis. Radiomics, which aims to extract imaging biomarkers that are imperceptible to the naked eye, also holds promise for characterizing lung nodules and predicting clinically relevant outcomes in cases of lung cancer. Deep learning for the detection and classification of lung nodules in chest radiographs and CT scans is gradually becoming integrated into routine clinical practice, while other deep learning and radiomics approaches are not yet widely applied in daily practice. Sustained efforts will be necessary to facilitate the clinical adoption of deep learning and radiomics, ultimately enhancing their contribution to patient management and outcomes.