Lung-AttNet: An Attention Mechanism-Based CNN Architecture for Lung Cancer Detection With Federated Learning

Lung cancer is one of the leading causes of mortality worldwide, with early detection playing a critical role in improving patient survival rates. This study presents Lung-AttNet, a novel Attention Mechanism-Based Convolutional Neural Network (CNN) architecture for the detection and classification of lung cancer images into three categories: Benign cases, Malignant cases, and Normal cases. The proposed model leverages an attention mechanism to focus on the most relevant features in medical images, improving the detection accuracy by efficiently highlighting important regions. To ensure robust model performance and generalization, we employ 5-fold cross-validation and integrate Explainable AI (XAI) techniques such as LIME for model interpretability, allowing us to understand and explain the decision-making process of the model. Additionally, Federated Learning is applied to enable decentralized model training across multiple datasets, addressing concerns related to data privacy and ensuring that the model can be trained on data from multiple sources without compromising confidentiality. The results demonstrate the efficacy of Lung-AttNet in achieving high classification accuracy, making it a promising tool for clinical applications in lung cancer detection. This work contributes to advancing the use of deep learning in medical diagnostics, offering a reliable and explainable solution for early lung cancer detection.

Keywords:

Lung Cancer Detection, Attention Mechanism, Convolutional Neural Network (CNN), Federated Learning, Explainable AI (XAI), LIME, 5-fold Cross-validation, Medical Imaging, Malignant Cases, Benign Cases, Normal Cases, Deep Learning.