The February 2023 issue of IEEE Spectrum is here!

Close bar

AI Predicts Asymptomatic Carriers of COVID-19

Machine learning identifies superspreaders and ranks source of infection among a moving crowd

2 min read
People wearing face masks visit Chunxi Road, the busiest commercial pedestrian Street in Chengdu, China, on November 28, 2020.
Photo: Gao Han/VCG/Getty Images

One of COVID-19’s nastiest tricks is the way it can infect someone and not cause any symptoms. This allows the virus to proliferate under the radar of contact tracers. But new artificial intelligence could help track down these silent carriers. 

In a paper published Friday in the journal Scientific Reports, researchers at Synergies Intelligent Systems and Universität Hamburg describe a machine learning algorithm that can identify people in a moving crowd who are most likely asymptomatic carriers of the virus that causes COVID-19. The algorithm makes these predictions based on the GPS-tracked movement of people in a city environment, and known cases of infection. 

Such precision contact tracing could reduce the number of people needlessly quarantined by lockdowns and conventional contact tracing, according to the authors of the paper. “With this type of technology, we can quarantine a very small fraction of people—just three to five percent—and pretty effectively reduce the effect of the disease,” says Michael Chang, co-founder of Synergies.

The tool could also be used to identify superspreaders: people infected with the virus who transmit it to a disproportionately high number of people. This could aid public health leaders in prioritizing where to concentrate vaccine supplies to head off an outbreak, Chang says.  

One limitation of the tool is that its accuracy depends on people willingly using a GPS-based app on their phones. Such an app would track their location to an accuracy of one meter, and log any positive COVID-19 test results. 

“The mobile data is also collectible from cellular data,” says Jianwei Zhang, a professor at Universität Hamburg and an author of the paper. “If we could use data other than GPS to build interactive relations between people—for example by using cameras or indoor localization methods—this would boost the [algorithm’s] prediction accuracy.”

A number of researchers developed digital contact tracing apps early on in the pandemic. Such software was used extensively in East Asia, drawing criticism over privacy rights of citizens. The technology was not widely adopted in many other regions of the world. 

The new machine learning technology boosts the accuracy of digital contact tracing by factoring in asymptomatic transmission. The authors call their system “continuous learning and inference of individual probability” or CLIIP. It’s based on a gradient boost ensemble learning tree model. The model is updated with real world data using individual directed graph, or IDG, which is the direction for recording the interactions between individuals at a certain time. 

The algorithm also integrates real world data such as credit card transactions, public transport data, and building cameras to determine how long and how often people came into contact, and how much physical distance was between themThe real world data came from a city in China that the authors would not disclose.

The algorithm can also factor in mask wearing, hand washing and health. “Such information can be given by a user via an interactive interface or automatically extracted using further sensors,” says Zhang. “This would make the CLIIP predictions more precise.”  

The system labels people based on their disease status, such as susceptible, quarantined, exposed, infected, hospitalized and recovered. “By using these labels, we are using our interactive model to learn what would happen between one guy and his surroundings,” says Shangching Liu, an AI engineer at Synergies who co-authored the paper. Synergies Intelligent Systems is an AI company that builds augmented analytics technologies that decrease the barriers to AI.

The Conversation (0)
Illustration showing an astronaut performing mechanical repairs to a satellite uses two extra mechanical arms that project from a backpack.

Extra limbs, controlled by wearable electrode patches that read and interpret neural signals from the user, could have innumerable uses, such as assisting on spacewalk missions to repair satellites.

Chris Philpot

What could you do with an extra limb? Consider a surgeon performing a delicate operation, one that needs her expertise and steady hands—all three of them. As her two biological hands manipulate surgical instruments, a third robotic limb that’s attached to her torso plays a supporting role. Or picture a construction worker who is thankful for his extra robotic hand as it braces the heavy beam he’s fastening into place with his other two hands. Imagine wearing an exoskeleton that would let you handle multiple objects simultaneously, like Spiderman’s Dr. Octopus. Or contemplate the out-there music a composer could write for a pianist who has 12 fingers to spread across the keyboard.

Such scenarios may seem like science fiction, but recent progress in robotics and neuroscience makes extra robotic limbs conceivable with today’s technology. Our research groups at Imperial College London and the University of Freiburg, in Germany, together with partners in the European project NIMA, are now working to figure out whether such augmentation can be realized in practice to extend human abilities. The main questions we’re tackling involve both neuroscience and neurotechnology: Is the human brain capable of controlling additional body parts as effectively as it controls biological parts? And if so, what neural signals can be used for this control?

Keep Reading ↓Show less
{"imageShortcodeIds":[]}