Emnet: A System for Privacy-preserving Statistical Computation on Distributed Health Data
Year: 2015
Type: dissertation
Abstract: Motivation: Despite its enormous benefits, EHR data reuse is limited because of multi-dimensional challenges where privacy comes on the forefront. Recently various privacy-preserving statistical computation tools have emerged. However, they have limited privacy guarantee and use ad-hoc techniques for privacy-preserving computation of statistical functions. \nPurpose: The purpose of this thesis is to develop a system that enables to compute a wide \nvariety of statistical functions on distributed EHRs, while preserving the privacy of patients \nand health institutions. \nMaterials and Methods: Systematic literature review of privacy-preserving techniques for \nhealth data reuse was performed to understand the state-of-the-art. The result of the review \nand meetings with users were used as sources of requirements. Agile methodology was used \nfor implementation of a prototype system called Emnet.\nEmnet uses openEHR-based EHRs as common data model to achieve interoperability among \nhealth institutions. We have prepared test openEHR data sets and a virtual environment that \nsimulates the real working environment for testing. \nResult: We have developed and tested privacy-preserving techniques for research data set \npreparation and statistical computation. The research eligibility criteria and required \nattributes are expressed as a computable query using Archetype Query Language (AQL), and \neach health institution executes the query and locally stores the resulting data set. The data \nsets are physically distributed across the health institutions, yet they collectively make the \nresearch data set, which we call Virtual Dataset.\nStatistical computations on the Virtual Dataset are performed using two main techniques, (1) \ndecomposition of statistical functions into summation forms and described as a computation \ngraph; and (2) secure summation protocols.\nConclusion: The developed techniques enable statistical computation on distributed health \ndata, while preserving the privacy of patients and health institutions. Currently, mean, \nvariance, Standard Deviation, Covariance and Pearson’s r are implemented in Emnet. \nHowever, the techniques are generic to implement more statistical functions, as long as they \ncan be decomposed into summation forms. The work presented in this thesis contributes for \nadvancement of privacy-preserving health data reuse. It is also relevant to other domains \nwhere they have similar requirements as health care.
Author: Hailemichael, Meskerem Asfaw
Institution: UiT The Arctic University of Norway
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2470344164