A Hybrid Quantum Resistant Secure Framework for Federated Healthcare Cyber-Physical Systems

The study proposes framework that integrates Quantum Key Distribution (QKD) with Advanced Encryption Standard (AES-256) to enhance the security of distributed healthcare analytics. The proposed model enables end to end encryption by allowing decentralised training across multiple healthcare institutions. To improve the model’s adaptability across diverse clinical environments both manual and automated feature selection techniques are incorporated under federated learning settings. The framework further integrates access control mechanisms within the federated learning update pipeline to address multi layered security requirements in healthcare systems. The evaluation of the framework has been measured through simulations conducted on the University of Southern Queensland (UniSQ) High Performance Computing (HPC) cluster. The implementations are mostly based on the python libraries for AES encryption, key generation, alongside federated logistic regression, and mutual information based feature selection. Experimental results demonstrate that the proposed hybrid framework achieves strong data confidentiality while maintaining model accuracy. The system effectively mitigates data breach risks with minimal computational overhead. Overall, the proposed framework provides a scalable and practical solution for quantum resilient healthcare analytics. This work establishes a foundation for future research in quantum safe federated learning techniques for healthcare systems.