Lung Cancer Detection System with XAI

Document Type : Original papers

Authors

1 Department of Software Engineering, College of Computing and Information Technology, Arab Academy for Science, Technology &Maritime Transport, Aswan, Egypt

2 Department of Computer Science, College of Computing and Information Technology, Arab Academy for Science, Technology &Maritime Transport, Aswan, Egypt

3 Department of Electronics and Communication, College of Engineering and Technology, Arab Academy for Science, Technology &Maritime Transport, Aswan, Egypt

Abstract

Lung cancer detection using deep learning significantly improves patient outcomes by enabling timely treatment and diagnosis that crucial for survival. Deep learning techniques introduce high efficiency by allowing the analysis of complex medical data, making them valuable in enhancing diagnostic precision. This study presents a proposed system leveraging artificial intelligence (AI) algorithms for the early detection of lung cancer, addressing the significant global mortality associated with the disease. In addition, the system utilizes nail image analysis for curvature assessment and user-provided data such as smoking history and genetic predisposition. A complementary dataset was created to enable rapid and accurate detection. Advanced AI models, Explainable AI (XAI) using the SHAP model, were employed to extract important details from the data, enhancing detection accuracy. The system also identifies additional diagnostic markers, such as clubbing nails, a key indicator of lung cancer. The results show the superiority of the proposed system with a detection accuracy of 96.0% and a loss rate of 0.07.

Keywords

Main Subjects

Files

Share

How to cite

Statistics