e-PrescripChain: A Secure and Patient- Centric Electronic Prescription System Using Blockchain and Federated Learning

This paper presents <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e-PrescripChain</i>, a novel blockchain-based electronic prescription system enhanced with federated learning (FL) to address critical security, privacy, and efficiency challenges in traditional e-prescription platforms. While existing systems like New Zealand’s NZePS have improved prescription accuracy, centralized architectures remain vulnerable to fraud and data breaches, as evidenced by real-world incidents. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e-PrescripChain</i> leverages Ethereum smart contracts for tamper-proof prescription management, IPFS for decentralised storage of sensitive data, and FL for privacy-preserving collaborative fraud detection across healthcare institutions. Experimental results demonstrate the system’s practicality: AES-256 encryption handles 500 KB prescriptions in under 500 ms, while IPFS ensures reliable data retrieval (80–100 ms access times). By combining blockchain’s immutability with FL’s distributed intelligence, the framework achieves compliance with healthcare regulations (e.g., NZ Privacy Act 2020), mitigates single points of failure, and enables real-time prescription tracking. This work advances secure e-prescription systems by addressing the limitations of centralised models through a scalable, patient-centric approach.