In DepthCOVID-19

AI systems aim to sniff out coronavirus outbreaks

See allHide authors and affiliations

Science  22 May 2020:
Vol. 368, Issue 6493, pp. 810-811
DOI: 10.1126/science.368.6493.810

Science's COVID-19 coverage is supported by the Pulitzer Center.

The international alarm about the COVID-19 pandemic was sounded first not by a human, but by a computer. HealthMap, a website run by Boston Children's Hospital, uses artificial intelligence (AI) to scan social media, news reports, internet search queries, and other data for signs of disease outbreaks. On 30 December 2019, it spotted a news report of a new type of pneumonia in Wuhan, China, and issued a one-line email bulletin that seven people were in critical condition, rating the urgency at three on a scale of five.

Humans weren't far behind. Colleagues in Taiwan had already alerted Marjorie Pollack, a medical epidemiologist in New York City, to social media chatter in China that reminded her of the 2003 outbreak of severe acute respiratory syndrome (SARS), which spread to dozens of countries and killed 774. “It fit all of the been there, done that déjà vu for SARS,” Pollack says. Less than 1 hour after the HealthMap alert, she posted a more detailed notice to ProMED, a list server with 85,000 subscribers for which she is a deputy editor.

But the early alarm from HealthMap underscores the potential of AI, or machine learning, to keep watch for contagion. As the COVID-19 pandemic continues to spread, AI researchers are teaming with tech companies to build automated tracking systems that will mine social media, news reports, and public health data for signs of new outbreaks. AI is no substitute for traditional public health monitoring, cautions Matthew Biggerstaff, an epidemiologist with the U.S. Centers for Disease Control and Prevention (CDC). “This should be viewed as one tool in the toolbox,” he says. But it can fill a need, says Elad Yom-Tov, a Microsoft computer scientist who has worked with public health officials in the United Kingdom. “There's such a wealth of data, we will need some sort of tool … to me that tool is machine learning.”

Well before COVID-19 hit, CDC began an annual competition to most accurately predict the severity and spread of influenza across the United States. The competition receives dozens of entries each year; Biggerstaff says roughly half involve machine learning algorithms, which learn to spot correlations as they are “trained” on vast data sets. Roni Rosenfeld, a computer scientist at Carnegie Mellon University, and colleagues have won the contest five times with algorithms that mine data including Google searches, Twitter posts, Wikipedia page views, and visits to the CDC website.

Teams involved in the flu challenge have now pivoted to COVID-19. They are applying AI in two ways. One aims to spot the first signs of a new disease or outbreak, just as HealthMap did. That requires the algorithms to look for ill-defined signals in a sea of noise, a challenge on which a well-trained human may still hold the upper hand, Pollack says.

AI could also assess the current state of an epidemic—so-called now-casting. The Carnegie Mellon team aims to now-cast COVID-19 across the United States, using data collected through pop-up symptom surveys by Google and Facebook, Google search data, and other sources in order to predict local demand for intensive care beds and ventilators 4 weeks out, Rosenfeld says. “We're trying to develop a tool for policymakers so that they can fine-tune their social distancing restrictions to not overwhelm their hospital resources.”

Although automated, AI systems are still labor intensive, notes Rozita Dara, a computer scientist at the University of Guelph who has tracked avian influenza and is turning to COVID-19. “By the time you get to AI, it's the easy part,” she says. To train a program to scan Twitter, for example, researchers must first feed it examples of relevant tweets, painstakingly selected by hand, Dara says. AI may also struggle in a rapidly evolving pandemic, where correlations between online behavior and illness can shift.

Embedded Image

HealthMap uses artificial intelligence and data mining to spot disease outbreaks (colored dots) and issue alerts.


AI has misfired before. From 2009 to 2015, Google ran an effort called Google Flu Trends that mined search query data to track the U.S. prevalence of flu. At first the system did well, correctly predicting CDC tallies roughly 2 weeks ahead of time. But from 2011 to 2013, it overestimated flu prevalence, largely because researchers didn't retrain the system as people's search behavior evolved, Yom-Tov says. Searches for news reports about the flu were misinterpreted as signs of infection.

“I don't think it's an inherent problem,” Yom-Tov adds. “It's something that we've learned from.” He and colleagues from University College London recently posted a paper to the arXiv preprint server showing they could correct for that media-related bias.

Nations struggling to adequately test for the new coronavirus might be tempted to use automated surveillance instead. Biggerstaff says that would be a mistake. When the flu re-emerges this fall, he says, only testing will be able to distinguish outbreaks of it and COVID-19. But AI might help policymakers direct more testing to hot spots. “The hope is that you would actually have the two working together,” says John Brownstein, an epidemiologist at Boston Children's who co-founded HealthMap in 2006.

Some researchers question whether AI systems will be ready in time to help with COVID-19. “AI will not be as useful for COVID as it is for the next pandemic,” Dara says. Still, machine learning in epidemiology seems here to stay. Pollack, who sounded the alarm about COVID-19 the old-fashioned way, says she, too, is working on an AI program to help scan Twitter for mentions of the disease.

View Abstract

Stay Connected to Science

Navigate This Article