Lightweight Privacy-Preserving Medical Diagnosis in Edge Computing
K Manikanta, G Priyanka, S Saifulla, J Raviteja, V Sthuthikumar, S Saif Ali
Problem: Healthcare diagnosis systems face challenges in privacy preservation, real-time processing,
and scalability, especially when using machine learning models like XGBoost. Solution: This paper
proposes a streamlined privacy-preserving XGBoost model for healthcare diagnosis in edge computing
environments. The model ensures data confidentiality through homomorphic encryption and lightweight
computation for resource-constrained edge devices.
