If you’ve ever suspected you might have a fever but were unable to find a thermometer, you’re not alone. A fever is frequently cited as a symptom of COVID-19 and other viral infections, making a temperature check crucial for quick diagnoses and preventing the spread of the virus. Unfortunately, accurate at-home thermometers aren’t widely available, despite the rise of telehealth consultations.
There are several reasons for this. Thermometers can range in price from $15 to $300, and many people only need them a few times a year. During times of sudden demand, such as the early days of the COVID-19 pandemic, thermometers can sell out, leaving many people, especially those in under-resourced areas, without access to a vital medical device when they need it most.
To address this issue, a team of researchers at the University of Washington has developed a smartphone app called FeverPhone that transforms smartphones into thermometers without requiring new hardware. Instead, the app uses the phone’s touchscreen and existing battery temperature sensors to gather data that a machine learning model uses to estimate users’ core body temperatures. In a test on 37 patients in an emergency department, FeverPhone estimated core body temperatures with accuracy comparable to some consumer thermometers. The team’s findings were published on March 28 in Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies.
Joseph Breda, a UW doctoral student in the Paul G. Allen School of Computer Science & Engineering, said:
“In undergrad, I was doing research in a lab where we wanted to show that you could use the temperature sensor in a smartphone to measure air temperature. When I came to the UW, my adviser and I wondered how we could apply a similar technique for health. We decided to measure fever in an accessible way. The primary concern with temperature isn’t that it’s a difficult signal to measure; it’s just that people don’t have thermometers.”
The app is the first to use existing phone sensors and screens to estimate whether people have fevers. It needs more training data to be widely used, Breda said, but for doctors, the potential of such technology is exciting.
Dr. Mastafa Springston, a co-author on the study and a UW clinical instructor at the Department of Emergency Medicine in the UW School of Medicine, added:
“People come to the ER all the time saying, ‘I think I was running a fever.’ And that’s very different than saying ‘I was running a fever. In a wave of influenza, for instance, people running to the ER can take five days, or even a week sometimes. So if people were to share fever results with public health agencies through the app, similar to how we signed up for COVID exposure warnings, this earlier sign could help us intervene much sooner.”
Did you know that clinical-grade thermometers use tiny sensors called thermistors to estimate body temperature? It turns out that off-the-shelf smartphones also contain thermistors, which are primarily used to monitor the temperature of the battery. Researchers at the University of Washington discovered that they could use these sensors to track heat transfer between a person and their phone. By measuring skin-to-phone contact and gauging the air temperature and the rise in heat when the phone touches a body, they were able to estimate body temperature.
To test their theory, the researchers conducted experiments in a lab. They used a sous-vide machine to heat a plastic bag of water to simulate a warm forehead, and then pressed phone screens against the bag. They tested three phone models and added accessories, such as a screen protector and a case, to account for variations in circumstances. They also changed the pressure on the phone.
Using the data from their experiments, the researchers trained a machine learning model to estimate body temperature. By tracking how quickly the phone heats up and using touchscreen data to determine how much of the heat comes from a person touching it, they were able to calibrate the model to account for variations in phone accessories.
The researchers then tested their app, called FeverPhone, on people. They conducted a clinical trial at the UW School of Medicine’s Emergency Department, where they compared FeverPhone’s temperature estimates against readings from an oral thermometer. They recruited 37 participants, 16 of whom had at least a mild fever. Participants held the phones with forefingers and thumbs touching the corner edges to reduce heat from their hands being sensed. Then, they pressed the touchscreen against their foreheads for about 90 seconds. FeverPhone estimated patient core body temperatures with an average error of about 0.41 degrees Fahrenheit (0.23 degrees Celsius), which is within the clinically acceptable range of 0.5 C.
Breda said,
“We started with smartphones since they’re ubiquitous and easy to get data from. I am already working on seeing if we can get a similar signal with a smartwatch. What’s nice, because watches are much smaller, is their temperature will change more quickly. So you could imagine having a user put a Fitbit to their forehead and measure in 10 seconds whether they have a fever or not.”
