Standardizing PHR: New File Format and Technical Challenges for Personal Data Management

Details

This white paper from the Zhejiang Digital Medical Health Technology Institute focuses on building Personal Health Records (PHR) through its 'Danggui' project. Unlike traditional government-led records, PHR integrates personal data from outside hospitals—such as clinic reports, physical examination results, and wearable device data—from an individual's perspective. It emphasizes that with AI and IoT technologies, PHR is evolving beyond static storage into a 'smart health engine,' capable of predicting health risks and assisting clinical care. The paper identifies three major obstacles to PHR adoption: first, ambiguous data ownership, addressed by establishing a rights structure through 'data triple custody' (holding right, usage right, operation right). Second, the lack of semantic interoperability among multi-source heterogeneous data, which proposes standardizing unstructured data using HL7 FHIR standards combined with Large Language Models (LLM). Third, privacy and data leakage risks are mitigated by building a trustworthy data space utilizing DID, privacy computing, and blockchain. The core achievement is the introduction of the proprietary format, '.phr'. This format uses a ZIP container + XML architecture to securely encapsulate diverse health files like PDFs and images within a fixed structure, ensuring traceability and security. This standard aims to provide the technical foundation for individuals to autonomously manage and utilize their data, establishing a new norm for compliant medical data flow.