VHA launches analytics challenge to predict COVID-19 outcomes
The Veterans Health Administration, part of the U.S. Department of Veterans Affairs, is calling for members of the public to develop machine learning and artificial intelligence models to predict health outcomes related to the novel coronavirus.
The challenge opened at the beginning of June, and VHA representatives say 84 participants have registered so far.
WHY IT MATTERS
COVID-19 has a particularly deadly effect on those over the age of 65 and people with underlying conditions, such as heart disease or diabetes. Because many veterans fall into those categories, the VA has moved to implement several facets of a response plan – including providing outpatient care through telehealth, offering testing for the virus and implementing outreach.
Now, the VHA is asking scientists and analysts to use synthetic veteran health records to create and test computational models predicting COVID-19 related health outcomes in veterans.
The outcomes include COVID-19 status, length of hospitalization and time in the ICU, and mortality.
"We encourage participants to use demographic data and the presence of comorbidities when developing their model to help precisionFDA and the VHA Innovation Ecosystem better understand how race, ethnicity, age, and comorbidities can affect the progression of COVID-19," said the VHA in a description of the challenge on the department's site.
"AI will be used to mine the data to look for predictive factors, including analyzing predictive value of each factor to then develop operation tools to guide clinical decisions within the VHA," said VHA Senior Innovation Fellow Amanda Purnell.
Challengers have until July 3 to enter. The success for the models will be determined, according to the VHA, by combining the ranks of model performance across all predicted outcomes.
Top performers may be invited to contribute to a manuscript, present at a conference or continue developing tools through the VHA Innovation Ecosystem.
THE LARGER TREND
Researchers have begun developing multiple AI models to aid in early detection and treatment. Last week, the U.S. Food and Drug Administration issued an emergency use authorization for a COVID-19 predictive screening tool from CLEW for use in the ICU.
"Healthcare providers need more than simple analytics. Systems need to integrate into the provider's workflow, offering ease of use and actionable data," said CLEW CEO Gal Salomon in a statement.
And in a Nature Medicine study last month, scientists at Mount Sinai Health System used AI alongside imaging and clinical data to diagnose patients with COVID-19 at a rate comparable to – and in some cases better than – experienced radiologists.
ON THE RECORD
"The challenge originated from an interest on the VHA to better understand underlying risk factors, including impacts of current care and contextual variables to provide better care," said Purnell.
Kat Jercich is senior editor of Healthcare IT News.
Healthcare IT News is a HIMSS Media publication.