ORNL and VA use security framework on Summit for health research
A team from Oak Ridge National Laboratory (ORNL) and the US Department of Veterans Affairs (VA) used the CITADEL security framework to securely transfer and analyze veterans’ health records on ORNL’s Summit, an IBM AC922 supercomputer housed at the Oak Ridge Leadership Computing Facility, a US Department of Energy (DOE) Office of Science User Facility. The success of this project will pave the way for future efforts and potentially life-changing research.
Led by Research Scientist Ioana Danciu at ORNL, the project is part of a DOE-VA partnership known as MVP-CHAMPION (Computational Health Analytics for Medical Precision to Improve Outcomes Now). The VA’s Million Veteran Program (MVP) collects medical data from veteran volunteers and maintains the largest biobank linked to electronic health records in the world. Partnering with ORNL, the VA can analyze this extensive and complex dataset by leveraging the lab’s high-performance computing (HPC) expertise and capabilities (e.g., Summit, CITADEL). The framework, which was unveiled last year, allows researchers to work with sensitive information, such as protected health data, on ORNL’s HPC systems.
Kelly Cho, director of MVP Phenomics, commented on the novelty of the current activities, “The current project is using Summit to build a comprehensive knowledge network of correlations among millions of parameters and concepts extracted from the VA electronic health record data. The complexity and size of the dataset means this analysis is only possible with the help of supercomputing.”
“From the knowledge discovery infrastructure, or KDI, we establish a secure tunnel onto Summit with CITADEL, perform our computation, and then transfer the data back,” explains Danciu.
“As the first users to test software development in the framework, working with the team at the National Center for Computational Sciences and with the KDI team, we enabled new capabilities in CITADEL for a fast and secure software development workflow. The complexity of adapting software for a supercomputer requires many iterations, and a fast workflow is needed,” says George Ostrouchov, distinguished computational statistician on the ORNL team.
In addition to Danciu and Ostrouchov, the ORNL team also consists of Everett Rush, research scientist; Ian Goethert, data engineer; Ryan Tipton, Linux systems engineer; Franciel Linares, R&D Systems Engineering team lead; and Jeremy Cohen, Scalable Protected Data program manager. The ORNL team worked closely with MVP Core staff from VA Boston and Harvard, including Kelly Cho, Tianxi Cai, Katherine Liao, Lauren Costa, Rachel Matty, Anne Ho, Vidul Ayakulangara Panickan, and Doudou Zhou. Together, they developed the methodology and optimized the software to run on Summit.
“This has been a joint accomplishment between the ORNL team and the VA team,” says Danciu. “It has been a truly collaborative effort.”
The success of this project has opened the door for further research using data from the MVP: the next cycle of VA-DOE projects will use data that spans 2 decades and will leverage Summit and CITADEL to study diseases ranging from cancer to mental health.
“We are trailblazers,” says Danciu.
UT-Battelle LLC manages Oak Ridge National Laboratory for DOE’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. DOE’s Office of Science is working to address some of the most pressing challenges of our time. For more information, visit https://energy.gov/science.