Conference: ICCES 2026
Publisher: ACM ICPS (Web of Science indexed)

This paper addresses privacy and governance challenges in healthcare AI by introducing an explainable federated learning framework. The approach enables collaborative model training across institutions without sharing raw patient data, while maintaining interpretability.

Key Contributions:

• Designed a federated learning architecture for diabetes prediction
• Preserved patient privacy while maintaining strong predictive performance
• Integrated explainable AI to ensure transparency in distributed models
• Aligned with GDPR-aware and ethically responsible AI principles
• This research highlights how privacy, performance, and explain ability can coexist in real-world healthcare systems.