Enhanced Attention-based Network for Lung Cancer Detection from 2D CT Scans

Abstract

Lung cancer remains a leading cause of cancer mortality, emphasizing the importance of early and accurate diagnosis. This study proposes an attention- based CNN model to enhance lung cancer classification from CT scans. The attention mechanism improves the model’s focus on critical regions, boosting diagnostic accuracy. Experiments on the IQ-OTH/NCCD and Lung Cancer Type datasets achieved classification accuracies of 99.65% and 98.3%, respectively, demonstrating significant performance improvements in distinguishing between cancer types and cases