Designing a secure rare disease data-sharing platform leveraging electronic health records: a system architecture and simulation study

The persistent fragmentation of rare disease data significantly hinders medical research and orphan drug development. This study aimed to design and technically validate a secure data-sharing platform capable of dismantling institutional data silos by activating underutilized electronic health records (EHRs). A multi-tier platform architecture was developed to integrate EHR data, incorporating the domain-adapted Rare-RoBERTa model and ORPHA-codes for clinical semantic standardization. To assess technical feasibility, a prototype was implemented on the Hyperledger Fabric blockchain to facilitate zero-trust, audited data exchange. Platform performance was evaluated using the Chinese Biomedical Language Understanding Evaluation (CBLUE) benchmark and synthetic clinical records, focusing on entity extraction accuracy, blockchain throughput, latency, and dynamic privacy-preserving de-identification. In simulated data governance evaluations, Rare-RoBERTa achieved an overall F1-score of 87.3% in named entity recognition (NER), outperforming general baseline models. The simulated blockchain network reached a peak write throughput of approximately 980 transactions per second (TPS) under 600 concurrent users, while the de-identification mechanism preserved research data utility with an information loss of 14.5% and a re-identification risk below 0.08%. These metrics represent simulated technical performance in a controlled environment and should not be interpreted as evidence of clinical effectiveness or nationwide scalability. The proposed architecture provides a preliminary, simulated proof-of-concept demonstrating the technical feasibility of rare disease data-sharing infrastructures. While these laboratory simulation metrics do not establish clinical effectiveness or operational scalability in live hospital environments, this study outlines a plausible architectural foundation to reactivate dormant clinical data and potentially support future health equity. Rigorous multi-center pilot trials using real-world heterogeneous EHRs remain essential prior to real-world clinical deployment, ultimately facilitating evidence-based policymaking and rare disease governance.

Authors

Institutions

Publication Details

Journal
Archives of Public Health
Published
2026-11-13
DOI
https://doi.org/10.1186/s13690-026-02138-0
Primary Topic
Electronic Health Records Systems
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Designing a secure rare disease data-sharing platform leveraging electronic health records: a system architecture and simulation study

Linyue Ma, Que Wu, Tiantian Dong
2 links
Archives of Public Health
Electronic Health Records Systems
article

Designing a secure rare disease data-sharing platform leveraging electronic health records: a system architecture and simulation study

Linyue Ma, Que Wu, Tiantian Dong
article en

Abstract

The persistent fragmentation of rare disease data significantly hinders medical research and orphan drug development. This study aimed to design and technically validate a secure data-sharing platform capable of dismantling institutional data silos by activating underutilized electronic health records (EHRs). A multi-tier platform architecture was developed to integrate EHR data, incorporating the domain-adapted Rare-RoBERTa model and ORPHA-codes for clinical semantic standardization. To assess technical feasibility, a prototype was implemented on the Hyperledger Fabric blockchain to facilitate zero-trust, audited data exchange. Platform performance was evaluated using the Chinese Biomedical Language Understanding Evaluation (CBLUE) benchmark and synthetic clinical records, focusing on entity extraction accuracy, blockchain throughput, latency, and dynamic privacy-preserving de-identification. In simulated data governance evaluations, Rare-RoBERTa achieved an overall F1-score of 87.3% in named entity recognition (NER), outperforming general baseline models. The simulated blockchain network reached a peak write throughput of approximately 980 transactions per second (TPS) under 600 concurrent users, while the de-identification mechanism preserved research data utility with an information loss of 14.5% and a re-identification risk below 0.08%. These metrics represent simulated technical performance in a controlled environment and should not be interpreted as evidence of clinical effectiveness or nationwide scalability. The proposed architecture provides a preliminary, simulated proof-of-concept demonstrating the technical feasibility of rare disease data-sharing infrastructures. While these laboratory simulation metrics do not establish clinical effectiveness or operational scalability in live hospital environments, this study outlines a plausible architectural foundation to reactivate dormant clinical data and potentially support future health equity. Rigorous multi-center pilot trials using real-world heterogeneous EHRs remain essential prior to real-world clinical deployment, ultimately facilitating evidence-based policymaking and rare disease governance.

Archives of Public Health
China Pharmaceutical University (CN)
Social Science Foundation of Jiangsu Province
Good health and well-being, Reduced inequalities
Electronic Health Records Systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon