Lung cancer is one of the most fatal types of cancer worldwide, and early detection plays a critical role in improving survivalrates. This project, Lung Insight, aims to enhance lung cancer detection from thoracic CT scans by integrating multiple datasets andapplying advanced image optimization techniques. The model will be based on deep learning, specifically leveraging ConvolutionalNeural Networks (CNN) to classify lung nodules as malignant or benign. Multi-dataset integration, including datasets like LIDC-IDRI,NSCLC, and LUNA16, will ensure model generalizability, while image enhancement techniques like Contrast Limited AdaptiveHistogram Equalization (CLAHE) will improve the clarity of CT scans for better feature extraction. The ultimate goal is to build arobust and efficient system that aids radiologists in early lung cancer diagnosis, improving patient outcomes through timely intervention.The system will also employ model optimization techniques like pruning and knowledge distillation to ensure faster inference anddeployment in real-time clinical environments. This study groups the existence of lung cancer in CT scan pictures and blood tests usinga framework for detecting the disease using a support vector machine and image processing. The focus of the work described in thisarticle is on the design and enhancement of a framework for determining lung cancer using CT scan pictures. In order to ensure improvedsurvival rates, it is crucial to arrange different tumour types. Lung cancer’s cycle of development is always being tested. The frameworkdetects the many stages of lung cancer, enabling experts to accurately and swiftly diagnose lung cancer from a wealth of information.Keywords: NLP, machine learning, Kaggle, CNN, CT scan pictures