Pulmonary nodules are pea-sized growths in the lungs that can be detected with imaging modalities such as computed tomography (CT). Accurate detection and classification of pulmonary nodules at an early stage is essential for the diagnosis and handling of various lung diseases, including lung cancer. Selecting the best of these numerous methods will be facilitated via the propagation of deep learning as a key enabling technology with the potential to enhance state-of-the-art medical imaging techniques such as computer-aided detection (CAD) systems that can classify and identify pulmonary nodules. This review discusses recent deep learning methods proposed for the analysis of pulmonary nodules. In the paper, critical approaches, datasets, and performance measures used in recent work are discussed, pointing to the impact of helpful preprocessing approaches. It also covers challenges like lack of labeled data, class imbalance, and interpretability of models and approaches to mitigate these challenges. Moreover, the review highlights emerging trends, such as the incorporation of attention mechanisms and multimodal data fusion, that have the potential to improve diagnostic accuracy. Finally, this survey intends to support investigators and industry experts in building effective and high performing deep learning platforms for pulmonary nodule detection and classification.