This investigation meets the urgent demand for efficient early diagnosis by presenting a thorough methodology for automated lung cancer categorization utilizing CT scans. The approach starts with the collection of a wide range of CT image databases and uses sophisticated image processing to improve visibility, such as pixel value normalization and Contrast Limited Adaptive Histogram Equalization (CLAHE). The process begins with the Histogram of Oriented Gradients (HoG) for feature extraction. Next, a Convolutional Neural Network (CNN)—more precisely, VGG Net—is fine-tuned to respond to the features of lung cancer. The optimized model is applied in the classification step, and strong generalization is guaranteed by performance assessment measures. Extensive research on the effects of different pre-processing processes complements the results. By providing insights into interpretability and reproducibility and advancing automated lung cancer detection, the research has potential uses for early diagnosis in clinical settings. This multidisciplinary strategy emphasizes how cutting-edge technologies can help solve difficult healthcare problems.