Healthcare Analytics, Population Health Management, Healthcare Big Data

Analytics in Action News

VA, DOE Launch Healthcare Big Data, Machine Learning Project

Researchers will combine the VA's vast healthcare big data assets with the Department of Energy's machine learning prowess to improve diagnosis and treatment of certain diseases.

Machine learning and healthcare big data analytics

Source: Thinkstock

By Jennifer Bresnick

- The Department of Veterans Affairs and the Department of Energy are launching their joint initiative to foster the development of healthcare big data analytics, machine learning, and artificial intelligence in order to support population health management and precision medicine.

The VA-DOE Big Data Science Initiative will be rooted in the DOE National Laboratory network, which is known worldwide for its supercomputing resources and expertise.  The five-year collaboration was announced in December of 2016 alongside a slew of other projects aimed at eradicating cancer and other serious conditions.

“Driving innovation through our national laboratories in ways that can improve Veterans’ health care is a remarkable opportunity,” said DOE Secretary Rick Perry. “I look forward to working together to shape this VA-DOE partnership.”

In addition to accessing the big data generated by the VA’s electronic health records, researchers will leverage the enormous wealth of genomic and clinical data collected through the Million Veterans Program (MVP) to study suicide risks, cancer, and heart disease.

The MVP databank has grown to include more than 560,000 volunteers who have contributed genomic, clinical, and lifestyle data to VA researchers. 

One of the first planned studies will focus on developing a more accurate and personalized suicide risk assessment score for veterans.  Researchers aim to implement more effective prevention and care coordination techniques that can supplement existing efforts to reduce self-harm among a highly vulnerable population.

A second project will explore variations in the development and survivability of prostate cancer, while a third study plans to tackle the intricacies of predicting and preventing cardiovascular disease.

“VA has developed unparalleled health data trend information from some 24 million Veterans who have used VA for health care over the past two decades,” said VA Secretary Dr. David J. Shulkin.

“We are partnering with DOE to use their high-performance computing capabilities to allow thousands of researchers access to this unprecedented data resource over time in a secure environment. The transformative science that will be developed through this partnership will improve health care for Veterans and all Americans.” 

The VA has taken a leading role in exploring how federal agencies can use their vast big data resources to improve health and pursue precision medicine.  The department is intimately involved in several aspects of the national Precision Medicine Initiative, including former Vice President Biden’s Cancer Moonshot.

In 2016, the VA committed more than $50 million to 250 research programs centered on precision medicine and advanced population health management, and has also formed collaborations with the National Cancer Institute and Department of Defense.

As part of these efforts, VA researchers have harnessed the potential of machine learning and cognitive computing through work with IBM Watson.  Speedier and more accurate cancer diagnosis is the goal of a VA-Watson partnership announced last summer, which put the supercomputer to work on sequencing tumors for thousands of veteran patients.

"The power of cognitive computing is its ability to ingest, understand and find patterns in massive volumes of disparate data – which is one of the fundamental barriers to precision medicine today," said John Kelly, senior vice president, IBM Research and Cognitive Solutions, at the time.

"In addition to helping advance clinical care, data and insights from Watson will also be shared with the research community, creating tremendous potential benefits for patients, researchers and society.”

Work with the DOE will likely follow similar lines, as the VA starts to dig into the millions of detailed data points available in the MVP databank alongside data assets from the Department of Defense, CMS, and the CDC’s National Death Index. 

Continue to site...