FeverPhone app turns phone into medical thermometer

Industry Updates Trending News Published 28th June 2023

A new app lets smartphones act as a thermometer for clinical purposes without the need for any extra hardware.

Modern smartphones already have temperature sensors to monitor the battery and internal temperatures.

FeverPhone app turns phone into medical thermometer

Lead researcher Joseph Breda had already worked on using a smartphone’s temperature sensor to measure air temperature.

On arriving at the University of Washington (UW) Paul G Allen School of Computer Science & Engineering, he decided to investigate the potential of using it to detect fever and measure human body temperature.

He said that the main difficulty in measuring temperatures was not that it was difficult, but that people did not tend to have ready access to thermometers.

Study co-author Dr Mastafa Springston said that people turning up at the emergency room with a confirmed temperature reading was different from them saying that they thought they were running a fever.

If results from the app could be shared with public health agencies, this could potentially speed up medical interventions.

System makes use of inbuilt smartphone thermistors

Smartphones contain thermistors or resistance thermometers, the same component utilised in clinical thermometer units.

While smartphone thermistors are primarily used to monitor battery temperature, the researchers realised that they could also be used to track heat transfer between the smartphone and the user.

The phone’s touchscreen is able to sense contact between skin and the device, while the thermistors can measure air temperature and gauge the rise in internal temperature during the contact.

To test the theory, the researchers used plastic bags of warmed water to simulate a forehead pressed against the screen.

They tested using three different smartphone models and other elements such as screen protectors and varying the pressure exerted.

They used the data obtained to train an AI machine learning model to track the complex interactions to estimate body temperature.

As the sensors are designed to monitor the temperature of the phone’s battery, the app tracks how quickly it heats up, combining this with the touchscreen data to work out how much of the increase stems from the person touching it.

In a clinical trial at the UW School of Medicine’s Emergency Department, the app was able to estimate core body temperatures – including people with mild fevers – with an average error of around 0.23C, which is well within the clinically acceptable range of 0.5C.

Today’s news was brought to you by TD SYNNEX – the UK’s number one solutions distributor.