Today, the US Department of Energy’s (DOE’s) Oak Ridge National Laboratory (ORNL) runs the largest mega-biobank in the world for the US Department of Veterans Affairs (VA). In a collaborative project with the VA, a team at the US Department of Energy’s (DOE’s) Oak Ridge National Laboratory (ORNL) has taken the medication possession ratio algorithm, which predicts risk of suicide, and engineered an expanded version that runs 300 times faster. The IBM AC922 Summit supercomputer at the Oak Ridge Leadership Computing Facility (OLCF) could cut the algorithms runtime down to seconds.
Oak Ridge National Laboratory located in Tennessee, near Knoxville, was founded in 1943 as part of the Manhattan Project and is the largest US DOE science and energy laboratory. Under UT Battelle LLC management it is conducting basic and applied research to deliver transformative solutions to compelling problems in energy and security.
UT-Battelle LLC was established in 2000 as a private not-for-profit company for the sole purpose of managing and operating the Oak Ridge National Laboratory for the US DOE. Formed as a 50-50 limited liability partnership between the University of Tennessee and Battelle Memorial Institute, UT-Battelle is the legal entity responsible delivering the DOE’s research mission at ORNL.
Until now, the medication possession ratio calculations have been limited in scope. The model has typically included only active psychotropic medications, such as narcotics or mental health medications, and covered a narrow class of the total veteran population in the Veterans Health Administration (VHA) database.
The ORNL team sped up the algorithm by applying advanced computing techniques for text mining and also increased its coverage to all current medications, as well as recent past prescriptions and all 9 million veterans in the database. Without the speedup, the expanded version of the model would have taken 75 hours to run. With the speedup, it runs in only 15 minutes, a 300-fold improvement.
Additionally, the VA hopes to use ORNL’s algorithm for analysis of the VA’s 3.5 billion clinical notes. Diagnostic codes and dates have been straightforward to input because they are structured but clinical notes have proved more difficult.
The VA program called REACH VET, or Recovery Engagement and Coordination for Health–Veterans Enhanced Treatment enables health care providers at 141 VA facilities to reach out to veterans who may be at risk of suicide. One of the VA’s goals is to implement the revised medication possession ratio algorithm into REACH VET, a task that will require a secure environment such as the one at ORNL.
“The medication possession ratio is one of our first examples of something that we know is a predictor of a variety of important outcomes.” … “If we can use it in our models like REACH VET, we will be able to better identify patients who might try to commit suicide.” – Jodie Trafton, director of the VA Program Evaluation and Resource Center in the VHA Office of Mental Health and Suicide Prevention
There’s no diagnostic code that says, ‘increased suicide risk’ or ‘job loss.’ If a patient reports a death in the family, a transient living situation, or social disconnectedness, we need to be able to identify those factors.” … “There’s a massive repository of these risk factors that hasn’t been used for predictive risk models.”
“Our ultimate goal would be to enable our operation to be a real-time alert system so that while physicians are interacting with a veteran, they can get live predictive risk alerts and intervene.” … “We don’t want veterans to walk into a clinic and get missed because someone hasn’t been specifically trained to recognize these symptoms. We never want it to be too late to reach someone.” – Edmon Begoli, principal investigator on the project and director of the Scalable Protected Data Facilities (SPDF), the National Center for Computational Sciences at ORNL
Sources
- https://www.ornl.gov/news/ornl-va-collaboration-targets-veteran-suicide-epidemic
- https://www.ornl.gov/content/solving-big-problems
- https://ut-battelle.org/about/
