IJPDS International Journal of Population Data Science IJPDS 2399-4908 Swansea University 10.23889/ijpds.v9i5.2898 9:5:406 menting Data Governance with Multi-Modal Privacy-Preserving Record Linkages between Restricted and Public Open Enclaves Gersing Kenneth 1 Michael Sam 1 Prior Fred 2 Saltz Joel 3 Moffitt Richard 4 Rogovin Sara 5 Phua Jasmin 5 Grannis Shaun 6 Amor Ben 7 Girvin Andrew 6 Baghal Ahmad 2 National Institutes of Health University of Arkansas Medical Sciences Stony Brook University Emory University Datavant Regenstrief Institute Palantir Technologies 18 09 2024 2024 9 5 2898 This work is licenced under a Creative Commons Attribution 4.0 International License. This article is available from the IJPDS website at: https://ijpds.org/article/view/2898

Data centralization is increasingly non-viable in the changing landscape of data privacy and data governance. New models of data sharing are needed, in particular when privacy-preserving record linkages (PPRL) are being used to connect data within restricted data enclaves and public open enclaves.

We share a privacy-preserving data-sharing infrastructure that employs a linkage honest broker implementation of PPRL connecting structured electronic health record data between the National Center for Advancing Translational Science (NCATS) National Clinical Cohort Collaborative (N3C), a restricted data enclave, with the National Cancer Institute (NCI) Cancer Imaging Archive (TCIA), a public open enclave.

In particular we will address:

Data governance models needed when linking data from restricted vs. public enclaves

Specific privacy-preserving consideration when linking to imaging data which contain specific structured data and imaging considerations

Privacy-preserving data aggregation when combining multi-modal data from restricted data enclaves with publicly discoverable imaging data