Lung cancer remains one of the most prevalent and deadly forms of cancer globally, with early detection significantly improving patient outcomes. In recent years, the advent of deep learning techniques, particularly 3D Convolutional Neural Networks (CNNs), has shown promise in automating the detection and classification of lung nodules from medical imaging data. This study proposes a novel approach to lung cancer detection using a 3D CNN architecture trained on volumetric computed tomography (CT) scans. The proposed method involves preprocessing the CT scans to extract informative features while reducing noise and artifacts. Subsequently, a 3D CNN architecture is designed to learn hierarchical representations directly from the volumetric data, enabling the model to capture spatial dependencies and contextual information crucial for accurate nodule detection. To enhance the model's performance, techniques such as data augmentation and transfer learning are employed to mitigate overfitting and leverage knowledge from pre-trained networks.