LUNG CANCER DETECTION USING CNN

Lung cancer remains a leading cause of cancer-related deaths worldwide, with early and accurate diagnosis being crucial for improving patient outcomes. Traditional diagnostic methods, such as imaging and biopsies, often face challenges in precision, subjectivity, and time efficiency. To address these limitations, this study proposes a deep learning-based framework leveraging Convolutional Neural Networks (CNNs) and ResNet architectures for lung cancer detection. Deep learning algorithms excel in extracting hierarchical features from complex data, making them particularly suitable for medical imaging analysis. CNNs are employed for their proven ability to identify spatial patterns and textures in lung scans, while ResNet overcomes the vanishing gradient problem through residual learning, enabling deeper and more robust networks. This combination allows the system to effectively differentiate between malignant and non-malignant cases. The proposed approach preprocesses lung CT images through noise reduction, adaptive histogram equalization, and segmentation to enhance image quality and focus on regions of interest. The CNN architecture extracts key features, and ResNet further refines these through deeper layers, learning intricate patterns indicative of cancer. The system is trained and evaluated on publicly available datasets, achieving a high degree of accuracy, sensitivity, and specificity in classifying lung cancer. This project results demonstrate the model’s capability to outperform traditional methods and existing deep learning models, with an accuracy of over 95% in detecting lung cancer. These findings underscore the potential of deep learning in revolutionizing cancer diagnostics, offering a fast, accurate, and non-invasive solution for early detection and improved treatment planning