University, UNC Health unveil SHIRE health care innovation platform

The Secure Health Informatics Research Environment will speed up the responsible use of artificial intelligence with clinical data.

UNC-Chapel Hill and UNC Health announced April 8 the launch of the Secure Health Informatics Research Environment, a secure, cloud-based platform designed to accelerate the responsible development of artificial intelligence using real-world clinical data.

SHIRE is the result of deep, sustained collaboration between UNC-Chapel Hill and UNC Health — built jointly and housed within UNC Health’s infrastructure — and represents a significant step forward in advancing data-driven health care innovation. The initiative is being jointly announced by Chancellor Lee H. Roberts and UNC Health CEO Dr. Cristy Page.

SHIRE provides researchers across UNC-Chapel Hill, the UNC School of Medicine and UNC Health with a powerful tool to develop and test advanced AI models using electronic health record data from UNC Health. Built with scalable cloud computing infrastructure and rigorous privacy safeguards, the platform enables intensive modeling on large-scale clinical datasets while maintaining the highest standards of patient data security.

Credentialed researchers can request and receive rigorously vetted data subsets drawn from a large clinical repository to bring real-world clinical evidence to bear on the treatments and interventions that matter most. Access to data in SHIRE is granted to Carolina and UNC Health researchers only through a structured review process, ensuring responsible stewardship of patient data at every step.

The design and implementation of SHIRE was led by the informatics and data science team at the North Carolina Translational and Clinical Sciences Institute in the medical school. Early work on the platform is already supporting research in areas such as precision oncology, rare disease identification and mental health. The platform is designed to grow with new external data sources, such as environmental and census data and other real-world signals. Researchers aim to build robust predictive models that can identify patients earlier, improve treatment targeting and support more personalized care.

"SHIRE reflects UNC's commitment to being at the forefront of AI, leading in AI research, and advancing responsible AI to improve patient care," Roberts said. "By bringing together real-world health data, cutting-edge computational infrastructure and strong governance, we are creating an environment where transformative health care innovations can be developed safely and effectively."

"At UNC Health and UNC School of Medicine, our goal is to be at the forefront of medical innovation and research that improves the health and well-being of patients across North Carolina and beyond," Page said. "UNC Health and the UNC School of Medicine are thrilled to have worked with the University on such an important initiative. We are excited to partner with researchers across UNC as they utilize this real-world data to accelerate the development of treatments and cures."

UNC-Chapel Hill's AI for Public Good Conference on April 13 marks the first public introduction of SHIRE and ORDR-D. The event will include presentations from Emily Pfaff, associate professor in the UNC School of Medicine's medicine department and co-director of informatics and data science at NC TraCS, and Brent Lamm, chief information officer for UNC Health, as well as a live Hackathon Burst with electronic health record data.

The platform operates alongside the de-identified Observational Medical Outcomes Partnership Research Data Repository. Known as ORDR(D), Carolina's secure analytical environment was built through a close collaboration between NC TraCS and the Renaissance Computing Institute and provides researchers with access to de-identified EHR data for research use.

While ORDR(D) supports broad discovery and analysis, SHIRE enables access to AI tools to assess unstructured clinical text and the potential integration of multimodal data sources. SHIRE and ORDR(D) combine to form a comprehensive research ecosystem designed to support the full life cycle of health care innovation — from data ingestion and model development to validation and clinical translation.

The launch of SHIRE opens new opportunities for Carolina, the UNC School of Medicine, and UNC Health to partner with technology companies, life sciences organizations and health care innovators in advancing AI-enabled solutions, with the shared infrastructure and governance to do so responsibly.

Originally posted at unc.edu.