Researchers are using artificial intelligence and machine learning to find better ways to protect people from this year's particularly aggressive flu season, according to NBC News.
So far this flu season, nearly 8,990 Americans were sent to the hospital and roughly 30 children have died, in part because this year's flu vaccine is only 30 percent effective.
How could AI better prepare the U.S. for flu seasons to come? Here are two ways.
1. Forecasting. Flu forecasts help researchers track outbreaks, tackle vaccine shortages and update the public, Roni Rosenfeld, PhD, a machine learning expert at Carnegie Mellon University, told NBC News. Dr. Rosenfeld and his team are applying machine learning techniques to sift through historical data on how the flu spread in past years.
"We make an underlying assumption that this year is going to be, in some senses, similar to a past year," Dr. Rosenfeld told NBC News. But he thinks the data could also offer insights into the flu's impact on individual counties or cities.
"When we track and forecast a flu in regions like the Southeast or New England, there's not a single epidemic going on," he said. "At the local level, the flu hits different cities and counties in different ways and at different times. They could really use a much more specific and customized assessment of what's going on."
2. Improving vaccines. Each year, the CDC characterizes about 2,000 strains of the influenza virus in order to identify which specific viruses are likely to dominate. This helps researchers develop targeted vaccines. However, because flu viruses mutate rapidly, the strain of the flu could change and leave the vaccine less effective.
Richard Webby, PhD, a virologist at St. Jude Children's Research Hospital in Memphis, Tenn., worked with a team of researchers to study how machine learning could examine the specific flu mutations that led to the 2009 swine flu outbreak. He said this study could be applied to the ordinary seasonal flu virus, too.
According to NBC News, scientists are using Dr. Webby's machine learning study to help pinpoint viruses' antigenic properties, or their molecular structures and how they affect the body's immune system.
"The most important piece of information that we have is antigenic information on these viruses that are circulating," Dr. Webby told NBC News. Scientists are attempting to "train machine learning algorithms to predict what impact various mutations might have on the antigenicity of the virus," meaning they are using the tech to find the right vaccine.