News

Study: Smartwatches could end the next pandemic

Everyday smartwatches are extremely accurate in detecting viral infection long before symptoms appear — now, research shows how they could help stop a pandemic before it even begins.
Two sets of graphs show infectiousness over time, one with and one without smartwatch alert, accompanied by icons.
An illustrative figure highlighting the impact of early detection via smartwatch on viral spread. The image shows how the model can capture both symptomatic (A and C) and asymptomatic cases (B and D).

Early detection of sickness is critical for preventing its spread — whether it’s COVID-19, influenza or the common cold. Yet, many illnesses are at their most contagious before people  even know they’re sick. Research shows that 44 percent of COVID-19 infections were spread several days before the sufferer came down with symptoms.

Person in a red-sleeved jacket standing on a pathway in front of a historic building with an arched entrance and palm trees.
Märt Vesinurm. Image: Robert Dehghan.

Now, researchers at Aalto University, Stanford University and Texas A&M, have released a study that models how smartwatches could provide a simple and effective way to dramatically minimise the unwitting spread of disease in pre-symptomatic or asymptomatic people. The research, published in prestigious US journal PNAS Nexus on March 4, reveals a powerful new tool for pandemic management.

‘Unlike during the pandemic, we now have concrete data on how pandemics develop, and how effective different measures are at curbing the spread. Add to this that wearable technology is now extremely effective when it comes to detecting the very early physiological signs of infection, and we are much better prepared,’ says Märt Vesinurm, from Aalto University in Finland. 

How an early ‘ping’ could flatten the curve

Numerous studies have proven the accuracy of wearable devices when it comes to recognising the physiological markers of various illnesses several days before the wearer themself. For example, everyday smartwatches can tell with 88 percent accuracy — from respiration and heart rate, skin temperature and other markers — if you’re coming down with COVID-19. And their accuracy rises to 90 percent accuracy for the flu, Vesinurm says. On average, people reduce social contact between 66-90 percent from the point when they realise they’re sick, even when not in a pandemic situation.

‘Even at the lower end of compliance, if people receive and act on an earlier warning by self-isolating, the impact is significant. Even just a 66-75 percent reduction in social contacts soon after detection by smartwatches — keeping in mind that that’s on a par with what you’d normally do if you had cold symptoms — can lead to a 40-65 percent decrease in disease transmission compared to someone isolating from the onset of symptoms,’ says Vesinurm. 

Notably, the research shows that higher compliance, such as that seen in a pandemic situation, could effectively stop a disease in its tracks. So, could smartwatches help us manage potential H5N1 (avian flu) epidemic? 

‘I see no reason why not,’ says Vesinurm. ‘As we gather more specific data about how different illnesses affect these measurements, there’s no reason we couldn’t distinguish between diseases, from bird flu and HIV to the common cold, especially when used in conjunction with advanced machine learning methods and other data from the user.’ 

Towards smarter pandemic policies

The team of researchers are the first to take real world data from numerous peer-reviewed epidemiological, biological and behavioural studies, drawing it together to mathematically model how infection spreads at a population level. Through this powerful combination of data and tech, Vesinurm foresees a future in which smartwatches change the way we deal with pandemics, at both individual and policy level. 

‘People are used to wearable devices and they’re likely to trust them. While they’re not yet diagnostic, they could help make unpopular approaches like masks, lockdowns and invasive testing more targeted and less of a blunt instrument,’ he says. An early warning from your smartwatch would be the cue to take other actions, like getting a PCR test, wearing a mask, or ‘not visiting your grandma’. The action would depend on the social context at the time, and an individual’s assessment of risk. 

Having this information not only empowers individuals to make choices that keep their loved ones or community safe, but it also gives decision-makers a powerful tool. 

‘It could be that governments find it most cost effective in a pandemic situation to provide every person who wants one with a smartwatch — although of course, this comes with its own ethical considerations,’ says Vesinurm. ‘Either way, with early detection literally at our fingertips, I see a lot of reason to hope.’

Märt Vesinurm

  • Updated:
  • Published:
Share
URL copied!

Read more news

Smiling couple in dark blue formal outfits, standing close together against a light blue studio background
Press releases Published:

Crown Princess Victoria and Prince Daniel to visit Aalto University in Espoo

The visit on Friday 21 August 2026 includes familiarising with quantum technology, space technology and ice research
Research & Art Published:

Comment on Finnish national Open Science policy

The National Open Science and Research Coordination invites comments from the scientific community on two Open Science policies.
Glitch art of flowing blue and red wave with Finnish text and Aalto University logo at bottom left
Research & Art, University Published:

New podcast explores Finnish contributions to artificial intelligence

Researchers Arno Solin ja Eric Malmi interview pioneers, entrepreneurs and experts about the AI field's history, disruptive consequences and the probability of humanity's destruction.
Hand holding a white tablet with text on screen, in front of a dark grid wall and bright light strips
Press releases, Research & Art Published:

AI model captures how humans read, paving the way to personalised text and better augmented reality

The new model could hold the key to developing highly customisable applications that adapt to different types of reader