Lung cancer is a significant public health concern, and early identification can improve patient outcomes. Advanced machine learning approaches can increase the precision of computer-aided diagnostic (CAD) systems that use medical pictures to diagnose diseases, such as CT scans, which can help find lung cancer. In this paper, we examine the usage of image analysis and CAD while also proposing an unique method for predicting lung cancer using Convolutional Neural Network (CNN) approaches. Our study utilized a dataset of lung CT scans from patients with and without lung cancer. The lung regions' important characteristics were extracted from the preprocessed pictures using segmentation and feature extraction techniques. After that, we used these variables to train a CNN model to identify the likelihood of lung cancer. Our paper demonstrates the potential of CNN techniques and image analysis in predicting lung cancer with high accuracy rates. The suggested method might be utilized to create more precise and efficient CAD systems for diagnosing lung cancer, possibly resulting in early identification and better patient outcomes.Further research is required to examine the clinical applications of this technique and to confirm these findings in bigger datasets.