Deep Learning-Based Computer-Aided Diagnosis System for Lung Cancer Detection from CT Images: A Novel Approach

Early detection of lung cancer significantly enhances survival rates. Computer-Aided Diagnosis (CAD) systems based on Computed Tomography (CT) images aim to improve detection accuracy and reduce interpretation time. This paper presents a novel deep learning-based CAD framework for automated lung nodule detection and classification. We employ advanced pre-processing, lung segmentation, feature extraction using 3D convolutional neural networks, and false positive reduction techniques. The system is evaluated on publicly available datasets achieving superior sensitivity and specificity compared to classical methods. Experimental results demonstrate the potential of the proposed approach for clinical lung cancer diagnostics.