New Facility Combines Legal and Computational Readiness to Advance Biomedicine
August 9, 2019 — Immediate response to severe biomedical events can reduce impairment and save lives. But treatment can often be delayed by inconclusive tests or protective policy constraints that govern public health information.
A new effort underway at the U.S. Department of Energy’s (DOE) Argonne National Laboratory will ease access to biomedical data, enable more collaborative biomedical research and employ techniques in artificial intelligence that lend themselves to quicker patient diagnoses.
The Argonne Biomedical Learning Enclave (ABLE) officially opened its new research space in Argonne’s Theory and Computing Sciences building. Its unique combination of capabilities and established relationships with the nation’s top biomedical research facilities will make ABLE a powerful resource for collaboration between clinicians and computational scientists.
Current research on early diagnosis and treatment of traumatic brain injuries (TBI), for example, results from a collaboration with the University of California San Francisco, Lawrence Livermore and Lawrence Berkeley National Laboratories.
Initial consults on TBI often recommend a CT scan that cannot always identify or differentiate among different subsets of TBI, like post-traumatic stress disorder and schizophrenia.
The ABLE collaborators use advanced machine learning and deep learning models — models trained to learn from data and progressively improve predictions about new data — to create more detailed images to reach diagnoses much faster.
“Within computerized tomography (CT) scans, there are features not visible to the human eye that artificial intelligence can learn from and use to generate an approximation of a magnetic resonance image,” says Ravi Madduri, a computer scientist in Argonne’s Data Science and Learning division. “So what typically takes a few weeks of analysis at a hospital can be reduced to a few hours using ABLE’s computing capabilities.”
ABLE researchers are teaming with researchers across the national laboratories and clinicians with the U.S. Department of Veterans Affairs (VA) to develop machine learning techniques similar to those applied to TBI to tackle areas of pressing importance to Veterans’ health.
Through a system called the Corporate Data Warehouse, the VA consolidates the electronic health records of nearly 23 million Veterans into a database, and through the VA Million Veteran Program (MVP), the VA is gathering genomic and lifestyle data from one million Veteran volunteers. In accordance with approvals from its Institutional Review Board, the VAis making this unprecedented data combination accessible to these authorized teams of ABLE, VA, and national laboratory researchers within the Oak Ridge National Laboratory’s Knowledge Discovery Infrastructure enclave, where a copy of the VA and MVP data reside. Within the records and the genomics data are clues to risk factors for suicide or certain types of cardiovascular disease, for example, that can be used to develop targeted treatments.
“Every person is unique and every situation is different, but the information required to pinpoint a treatment plan for a particular patient is all in the data,” says Madduri, ABLEresearcher and national laboratory lead for efforts in genomic imputation and cardiovascular disease. “With deep learning, we can analyze these data and make these particular drug discoveries or treatment plans happen at a much faster rate.”
In addition, ABLE has addressed the ability of researchers and clinicians to easily share and compare vital healthcare data, whether to speed up diagnoses and treatment or advance collaborative efforts.
ABLE has also laid the legal groundwork that will streamline access to a secure repository of public health information in compliance with a breadth of security measures — from the Health Insurance Portability and Accountability Act (HIPAA) to the Federal Information Security Management Act (FISMA), among many others.
“There is a lack of infrastructure in many biomedical research environments that keeps people from easily collaborating on problems,” adds Madduri. “ABLE’s infrastructure provides a secure location where a team of people can work together on large biomedical issues, even when they involve data that must be protected and treated with certain levels of sensitivity.”
In addition to its work with TBI, cardiovascular disease and suicide prevention, ABLEresearchers are providing outreach to clinicians from other leading health care institutions around the nation, including the National Cancer Institute, the National Institute of Allergy and Infectious Disease and the National Heart, Blood, and Lung Institute.
“Having our top researchers work together on previously unavailable medical data has the potential to accelerate biomedical research,” says Rick Stevens, associate laboratory director of the Computing, Environment and Life Sciences directorate. “We’re excited about the opportunity to dig into these records, work with clinicians and provide advanced analysis that could positively impact health outcomes around the country.”
Argonne National Laboratory seeks solutions to pressing national problems in science and technology. The nation’s first national laboratory, Argonne conducts leading-edge basic and applied scientific research in virtually every scientific discipline. Argonne researchers work closely with researchers from hundreds of companies, universities, and federal, state and municipal agencies to help them solve their specific problems, advance America’s scientific leadership and prepare the nation for a better future. With employees from more than 60 nations, Argonne is managed by UChicago Argonne, LLC for the U.S. Department of Energy’s Office of Science.
The U.S. Department of Energy’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit https://energy.gov/science.
Source: John Spizzirri, Argonne National Laboratry