Automated Lung Cancer Diagnosis using Convolutional Neural Networks

Abstract

Lung cancer is a leading cause of cancer-related mortality worldwide, and early detection is essential for improving patient outcomes. Traditional diagnostic methods rely heavily on radiologists interpreting chest CT scans, a process that is time-consuming and subject to inter-observer variability known as Medical Image Analysis. This study proposes a Convolutional Neural Network (CNN) framework for automated lung cancer diagnosis using CT images. The dataset was preprocessed through normalization and augmentation to enhance model robustness and generalization. The CNN model was optimized to classify images as cancerous or non-cancerous, with performance evaluated using accuracy, precision, recall, F1-score, and AUC. Experimental results demonstrate high classification accuracy, indicating the model’s potential as a Computer-Aided Diagnosis (CAD) tool. Grad-CAM visualization further highlights discriminative regions, improving interpretability. This automated system offers a reliable, efficient approach to support radiologists, reduce diagnostic workload, and enhance clinical decision-making.

Keywords: Convolutional Neural Network (CNN), CT scans, Computer-Aided Diagnosis, Medical Image Analysis.