Building a Secure Electronic Health Record Management System on Cost-Effective Devices Using Non-Fungible Tokens

Abstract

Our EHR Management System (EMS) empowers patients to control their electronic health records (EHR), en-hancing data privacy and access control. The system allows patients to carry their data as modular units on cost-effective, resource-constrained devices like Raspberry Pi, Beaglebone, and ESP32. We implemented EMS in two scenarios: one device per patient and one device shared among multiple patients. Using Blockchain-based Non-Fungible Tokens (NFTs), our system ensures secure access and authentication for authorized users. We evaluated our EMS by measuring delays in accessing data and verifying NFTs in a hospital scenario where two types of patients are scheduled to general and specialized doctors hourly. Despite using low-cost devices, scheduling delays were minimal. Among the tested scheduling techniques-Modified Queue-based, Reinforcement Learning (RL), and Deep Reinforcement Learning (Deep RL)-the Modified Queue-based method showed the least delay, proving efficient for our EMS.