Context-Aware Privacy Orchestration for Cloud-based Electronic Health Record Systems

Abstract

Cloud-based EHR systems have posed great privacy concerns when it comes to healthcare data processing, analysis, and sharing. Currently available privacy solutions for healthcare data are applied separately and lead to the development of rigid, non-adaptive privacy frameworks in healthcare. The contribution of this paper lies in its system-level view on context-aware privacy-preserving architectures for cloud-based EHR systems. A new taxonomy of privacy-preserving architectures is proposed based on their compositionality, adaptability, and complexity in terms of integration. Moreover, a new concept of dynamic privacy orchestration is proposed to provide flexible selection of privacy-preserving methods depending on the sensitivity of processed information, characteristics of data queries, and other system-specific constraints. The benefits and tradeoffs related to different privacy architectures in terms of their security guarantees, scalability, computational requirements, and output confidentiality will be discussed. Finally, the paper also proposes a comprehensive orchestration-driven privacy framework that can be used for lifecycle-oriented privacy protection within secure healthcare data processing environments.