One of the nation's leading multi-sector computational research partnerships is expanding its reach into big health data analytics.
The research project, with $600,000 in core funding from the National Science Foundation, is being undertaken through the University of Maryland-Baltimore County's UMBC Center for Hybrid Multicore Productivity Research (CHMPR) and several academic and industrial partners including Northrop Grumman Corporation. The latest round of funding, good for five years, expands an existing five-year grant.
While UMBC and Grumman have an established partnership through CHMPR in cybersecurity research, it isn't that big a jump into healthcare analytics, according to Morgan Crafts, technology manager at Northrop Grumman.
"The UMBC high-performance project really got its roots in the cyber community, in security," Crafts said. "For the first five years of CHMPR, most the analytical techniques and work we were doing were based on cybersecurity initiatives. But those same issues have direct implications for health. The techniques useful from the cyber side, such as the ability to find outliers, also apply directly to doing big data mining in health."
The project's research follows its name leveraging existing computational paradigms as well as cutting-edge architectures in multicore processing that can aggregate data not only from widely disparate sources in single domains, but also across them.
"We are analyzing EHR data, lab data, genomic data, and text data with text mining and natural language processing," Sam Shekar, M.D. Northrop Grumman's chief medical officer, said. "The value really comes from the interoperability of the analysis itself, the insights and findings you get from combining different sets of data and getting the insights you would not get if you just did individual data sets alone."
UMBC engineering and computer science professor Yelena Yesha, executive manager of CHMPR at the university, said the consortium's national resources will help it perform research in providing "near real-time" intelligence for clinicians on advanced computer architectures.
"Timing is of the essence given the short duration of patient-doctor encounters," Yesha said. "Information comes from multiple streams. It can come from sensors, genetics databases, environmental data, and evidence-based medicine databases. So integration is another big component of our goals. In real time, they should be able to integrate all these data feeds and come up with enough intelligence that will be meaningful and effective."
Crafts said he expects CHMPR research to also have implications on the macro level as well as the exam room, looking for ways to discern key indicators of outcomes or costs "things like tobacco use or alcohol use, or what are co-morbidity factors across diseases like diabetes and cardiovascular conditions? Those kinds of things can be learned now without having to have a clinical trial and can actually inform policy and direction and performance now. The impact doesn't have to wait for whatever future bedside activity might happen."
While $600,000 may not sound like an enormous amount of money to bootstrap such an ambitious research agenda, Yesha said the center, which includes academic resources at the University of California, San Diego and the University of Utah, as well as industrial partners such as Northrop Grumman, Intermountain Healthcare, and IBM, is also sufficiently funded by membership dues and other competitively-awarded grants. She also said the consortium's technology transfer process is consistent and well-defined, so breakthroughs can be disseminated and capitalized upon quickly.
Register or login for access to this item and much more
All Health Data Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access