The Department of Veterans Affairs (VA) and the Department of Energy (DOE) are partnering to drive technology innovation and transform health care delivery for Veterans. The partnership brings together VA’s unparalleled and vast array of healthcare and genomic data with DOE’s world class high performance computing (HPC), artificial intelligence and data analytics. By combining expertise, we can push the frontiers of data analytics, next-generation computing, precision health, genomic sciences, and health care delivery. This partnership supports:
- Innovation tied to design and development of DOE’s next generation supercomputing that will merge Big Data (BD), Artificial Intelligence (AI) and High-Performance Computing (HPC) as well as innovation in population science using complex health system and genomic data for knowledge generation.
- Better Healthcare via using supercomputing to inform when and how to treat our Veterans to improve outcomes and reduce cost.
- Better Science via a cadre of researchers and clinicians who specialize in healthcare with the DOE experts in HPC, AI &BD.
- Better Government via interagency collaborations bringing to bear the full capabilities and expertise within and public private partnerships.
The starting point for the DOE-VA partnership is MVP CHAMPION (Million Veterans Program Computational Health Analytics for Medical Precision to Improve Outcomes Now). Under MVP CHAMPION, VA and DOE will establish a scientific computing environment that will not only house and provision MVP data to researchers nationally, but also create a big data environment to foster transformational science.
In addition to the unparalleled health data on about 24 million veterans who have used the VA for healthcare over the last two decades, data will be obtained from the Department of Defense, Centers for Medicare and Medicaid Services, and the Center for Disease Control’s National Death Index. The Million Veteran Program (MVP), which has already enrolled over 560,000 Veterans, will enhance these rich health data resources with genetic and self-reported lifestyle and military exposure information making this among the most valuable health care data repositories in the world.
At present, combined DOE/VA teams, including scientists from ANL, LANL, LLNL and ORNL, have taken advantage of DOE’s international internet backbone that connects the laboratory enterprise – ESNet – and an existing protected health information (PHI) enclave at ORNL to configure the initial environment for scientific analyses. Several VA electronic health record (EHR) and MVP data assets have been moved to the secure enclave that will allow investigators access to the data and computing resources within an environment that will protect our Veterans’ data. Direct, high-speed networks between the VA and DOE facilities are expected to be established by June 2017.
On April 17-18, 2017, VA and DOE scientists, physicians, and leadership came together to develop technical roadmaps for how HPC can develop solutions to priority issues in caring for our Veterans. VA priorities that were identified that could deliver early impacts were: Suicide Prevention, Prostate Cancer and Cardiovascular Disease. Specifically:
- Patient specific analysis for Suicide Prevention: Suicide is the 10th leading cause of death in the US, and is significantly higher in the Veteran population, with 20-22 deaths per day. Efforts would improve identification of patients at risk for suicide through new patient-specific algorithms built to provide tailored and dynamic suicide risk scores for to bring the resources to each veteran at risk. Working closely with VA’s Office of Suicide Prevention, the tools would be used to create a clinical decision support system that assists VA clinicians in suicide prevention efforts, and helps to evaluate effectiveness of various prevention strategies.
- Help doctors discern lethal from non-lethal Prostate Cancer: Prostate cancer patients may undergo surgeries or other deleterious treatments without knowing if such treatments are actually necessary or effective, since there is no way of determining lethality a priori. The collaborative prostate cancer project will build classifiers for prostate cancer which could significantly aid doctors in distinguishing lethal from non-lethal prostate cancers. Reducing unnecessary treatments will provide an increased quality of life for patients and allow VA to focus resources where most effective.
- Enhanced prediction and diagnosis of Cardio-vascular Disease (CVD): CVD is the leading cause of death in US men and women - including Veterans- and the cost of care for CVD conditions is high. A collaborative CVD project would build new predictive tools that (1) identify improved sets of risk factors for specific types of CVD and (2) develop methods that will inform individualized drug therapies. The new tools will enhance prediction, diagnosis and management of major CVD subtypes in Veterans.
Crosscutting technology Advances: Cutting across the three projects are requirements for next generation AI and BD analytics. These requirements push development of DOE technologies in key areas including large scale data analytics, computer modeling, large scale machine learning and natural language processing. Success in developing these enabling technologies will have large impacts on DOE missions including science, energy, and national security.
Senior leadership at Intel, IBM, GE and NVidia, as well as others - important private sector companies for next generation computing, data analytics and health care- are taking note of this joint effort. This partnership has the potential to bring in the private sector to share risk, develop prototypes and accelerate advances.
Currently, the United States is the only country in the world with the opportunity to partner such a large scale health database with world leading data analytics capabilities. With this partnership, the US can drive cutting edge technology in next generation data analytics and improve quality of life for veterans and all Americans.