SecureEdge-MedChain: A Post-Quantum Blockchain and Federated Learning Framework for Real-Time Predictive Diagnostics...

Area of Science:

• Computer Science
• Cybersecurity
• Medical Informatics

Background:

• The Internet of Medical Things (IoMT) faces challenges in scalability, data confidentiality against quantum threats, and real-time privacy-preserving intelligence.
• Existing IoMT systems struggle to meet the demands of high-volume data processing and robust security in healthcare.

Purpose of the Study:

• Introduce Med-Q Ledger, a novel framework to address limitations in IoMT scalability, data security, and privacy.
• Enhance real-time patient monitoring and predictive diagnostics through secure and efficient data analytics.

Main Methods:

• Integrated a permissioned Hyperledger Fabric with Holochain DHT for scalability and transactional integrity.
• Incorporated post-quantum cryptography (PQC) using CRYSTALS-Di lithium and Kyber Key Encapsulation Mechanisms for data security.
• Utilized edge-based federated learning (FL) with autoencoders for privacy-preserving anomaly detection on encrypted gradients.

Main Results:

• Achieved high throughput (~3400 TPS) with low latency (~180 ms) and >95% anomaly detection rate.
• Demonstrated superior performance in predicting colostomy necessity in preterm infants with a 0.90 F1-score.
• Reported an 11% PQC overhead, 25% reduction in emergency surgeries, and 31% lower energy consumption compared to baselines.

Conclusions:

• Med-Q Ledger provides a secure, scalable, and privacy-preserving framework for IoMT analytics.
• The framework sets a new benchmark for next-generation healthcare deployments, improving patient outcomes and operational efficiency.