Detection of Lung Carcinoma using Volumetric Convolution (V-Net)

Lung cancer is one of the leading causes of cancer-related deaths due to its aggressive nature and late detection at advanced stages. Timely detection of lung cancer is critical for a patient’s life and is a significant challenge. In general, chest radiography (X-rays) and Computed Tomography (CT) scans are utilized to diagnose malignant nodules; however, the presence of benign nodules leads to inaccurate choices. In the early stages, benign and malignant nodules seem strikingly similar. This research presents an innovative deep learning-based model with multiple approaches for the accurate diagnosis of malignant nodules. We discussed the Volumetric Convolution (V-Net) architecture for lung nodule detection and classification using the Lung Nodule Analysis (LUNA-16) dataset in this study.