Secure and interpretable lung cancer prediction model using mapreduce private blockchain federated learning and XAI

Area of Science:

• Artificial Intelligence in Oncology
• Medical Informatics
• Computational Biology

Background:

• Lung cancer remains a leading cause of cancer mortality globally.
• Effective early detection is crucial for reducing lung cancer mortality rates.
• Existing predictive models face challenges including high computational costs, data privacy concerns, limited data sharing, and lack of interpretability.

Purpose of the Study:

• To develop a novel, integrated AI framework for accurate and secure lung cancer prediction.
• To overcome the limitations of conventional predictive models in handling large datasets and ensuring data privacy.
• To enhance the interpretability of AI models for clinical application.

Main Methods:

• Utilized MapReduce for efficient processing of large-scale lung cancer datasets.
• Implemented Private Blockchain for secure, immutable, and tamper-proof patient data management.
• Employed Federated Learning (FL) to enable collaborative model training across institutions without compromising patient privacy.
• Integrated Explainable Artificial Intelligence (XAI) to enhance model transparency and trustworthiness for clinicians.

Main Results:

• The proposed integrated framework achieved a high accuracy of 98.21% in lung cancer prediction.
• Demonstrated a significantly low miss rate of 1.79%.
• Outperformed previous approaches in terms of accuracy, privacy preservation, scalability, and interpretability.

Conclusions:

• The novel AI model offers a secure, interpretable, and scalable solution for lung cancer prediction.
• The integration of MapReduce, Private Blockchain, FL, and XAI sets a new benchmark for AI in healthcare.
• The model's high performance and enhanced features support improved clinical decision-making and patient outcomes.