Why we need it?

Health data reuse, which refers to the use of data for purposes other than those for which the data were initially collected, has enormous potential for the advancement in the science of medicine. Often reuse of data from multiple data sites is necessary especially for rare cases or monitoring quality of healthcare at different geographical level, which is the focus of Emnet. However, data reuse raises legitimate privacy concerns from different stakeholders that restricts the reuse of health data. There is a huge need for distributed data analysis techniques that protect the security and privacy of the people and organizations these data represent. Emnet is a tool that aims to address this need.

How does it work?

Emnet is a tool for privacy-preserving distributed statistical computation (PPDSC), also known as privacy-preserving distributed data mining (PPDDM), which is an emerging approach for processing data distributed across multiple data sources while protecting privacy. Emnet allows running statistical algorithms on confidential data divided across three or more different data sources without allowing any party to view the private data of another data source. The system reveals statistics generated from the combined data for a group of data sources, which does not reveal sensitive information about the inputs.

Emnet contains software components deployed at each health institution and a semi-trusted third party, denoted as a coordinator. The coordinator aids the system without learning private information and is trusted to follow protocol specifications. This is a standard security model known as the honest-but-curious adversarial model. The coordinator accepts statistical queries and jointly executes secure protocols with healthcare institutions without allowing any party to view the private data that healthcare institutions locally store.

In addition to the privacy protection provided by Emnet, it also allows healthcare institutions to maintain access control on their data (see featured image showing how Emnet allows institutions to approve participation in a study) such as whether or not their data is used for a specific purpose, who can analyze their data, and what data is available for secondary use. In general, the capabilities provided by privacy-preserving distributed data analysis tools like Emnet ease the obstacles for unlocking the potential of routinely collected data.