CANCER DETECTION SYSTEM

ABSTRACT Lung cancer is a potentially lethal illness, posing challenges in both detection and treatment. Despite ongoing research, the true cause of cancer and its complete treatment remain undiscovered. Early detection significantly improves treatment outcomes, yet current methodologies struggle to efficiently identify early-stage cancer. Leveraging image processing techniques, such as noise reduction, feature extraction, and medical history data comparison, offers promise in locating cancerous regions within the lung. However, manual segmentation tasks prove time-intensive for radiation oncologists. To address this challenge, we propose a comprehensive system utilizing Convolutional Neural Network (CNN) models such as 3D-CNN and VGG16 variants, for the early diagnosis of lung cancer. In this study, the Kaggle chest LDCT-scan images dataset was employed to identify lung cancer lesions. The system aims to enhance early cancer detection, thereby maximizing patient recovery and survival chances. By analyzing thousands of high-resolution lung scans, our model accurately identifies cancerous lesions, minimizing false positives and enabling prompt access to life-saving interventions. This approach not only enhances patient outcomes but also reduces the time constraints faced by radiologists, allowing for more meaningful patient interactions. The proposed system holds promise for revolutionizing lung cancer detection, offering potential significant impacts on public health and medical practices.