From Clusters to Clouds: Designing Secure, HIPAA-Compliant Data Science Ecosystems for Patient-Level Analytics

Presentation Type: Panel
Presentation Time: 02:00 PM - 03:15 PM

Primary Track: Data Science/Artificial Intelligence


Secure, HIPAA-compliant environments for analyzing patient data are now essential to clinical and translational research, but there is no single “right” way to design them. Institutions are making different choices: strengthening on-premises high-performance computing, building cloud platforms, or combining both. This panel compares three production ecosystems that support patient-level data science today: an on-premises secure research computing platform at UAB, an Amazon Web Services environment used by the University of Missouri, the coordinating center for the Greater Plains Collaborative (GPC) network, and the new Azure-based Secure Health Informatics Research Environment (SHIRE) at UNC Health. Panelists will describe how they configure secure workspaces, manage access to sensitive data, and support analytics and machine learning while protecting privacy. Through contrasting experiences and shared design questions, attendees will gain a practical understanding of technical and governance patterns they can adapt to build or improve their own secure data science ecosystems.

Moderator:
Abu Mosa, PhD, MS, FAMIA
University of Alabama at Birmingham

Speaker(s):
Matthew Dunlap, MSc
University of North Carolina at Chapel Hill

Ralph Zottola, PhD
The University of Alabama at Birmingham

Vasanthi Mandhadi, Masters
University of Missouri

Author(s):
Abu Mosa, PhD, MS, FAMIA - University of Alabama at Birmingham; Ralph Zottola, PhD - The University of Alabama at Birmingham; Vasanthi Mandhadi, Masters - University of Missouri; James McClay, MD - University of Missouri; Matthew Dunlap, MSc - University of North Carolina at Chapel Hill; Emily Pfaff, PhD, MS - UNC Chapel Hill School of Medicine;

---