Lung cancer is a concern in terms of health with a higher mortality rate than bone, prostate, and colon cancers combined each year. To address the need for early detection of diseases, we have developed a system that aims to improve cancer detection. Our system serves as a tool by utilizing cutting-edge deep learning models like CNN, ResNet50, VGG16, and Inceptionv3. During our study, we extensively evaluated these models using CT-scan images to accurately identify three main types of lung cancer: adenocarcinoma, small cell carcinoma, and large cell carcinoma. Our primary objective was to establish a deep learning-based model for efficient detection. This research represents the quest to develop methods for detecting lung cancer. Our system relies on various learning techniques to validate its impact on detection. The proposed model is VGG16. The VGG16 model is a 16-layered architecture with 13 convolution layers and contains max pool layers in between, and after these layers, we have fully connected layers followed by the SoftMax layer. Then, by evaluating the model on other metrics, the proposed model provides the highest accuracy of 95.81% with precision of 98.37%, recall of 96.37%, and Fl score of 97.36%. In this research, we have also evaluated other models such as CNN, ResNet50, Inceptionv3 and they achieved an accuracy of 92.01%, 93.23%, and 92.36% independently.