Smartphones could soon be used to detect type 2 diabetes, thanks to a team of engineers from the University of California, San Francisco (UCSF), who have recently revealed this promising potential. The novel research demonstrates a simple technique that doesn’t require additional tools other than the smartphones existing camera. Impressively, this technique is more than 80% accurate in detecting diabetes.
Geoffrey Tison, the co-senior author, said:
The ability to detect a condition like diabetes that has so many severe health consequences using a painless, smartphone-based test raises so many possibilities. The vision would be for a tool like this to assist in identifying people at higher risk of having diabetes, ultimately helping to decrease the prevalence of undiagnosed diabetes.
The method they’re using is based on photoplethysmography (PPG), which is a technique that detects blood volume changes by shining light into tissue. Most people know PPG as the ‘finger clamps’ that doctors use to measure blood oxygen levels and heart rate. In this study, published on Aug 17 in the journal Nature Medicine, the researchers hypothesized that smartphone cameras could capture PPG data to detect vascular damage caused by diabetes.
Robert Avram, the lead author on the new study, said:
Diabetes can be asymptomatic for a long period of time, making it much harder to diagnose. To date, non-invasive and widely-scalable tools to detect diabetes have been lacking, motivating us to develop this algorithm.
First, the team developed a deep-learning algorithm that could effectively pinpoint individuals with diabetes from those without it, by sorting through millions of PPG recordings. In total, the network examined 2.6 million PPG recordings from 53,870 subjects with diagnosed diabetes.
The researchers then tested out the algorithm’s ability to detect diabetes from smartphone PPG data from only three separate groups. The information is collected via the device’s flashlight and camera when the user applies their fingertip to the lens. The results saw the system accurately detecting diabetes in roughly 80% of subjects. However, when the algorithm was provided with primary patient data, including age and mass index, the predictive potential improved even further.
Jeffrey Olgin, an author on the new study, suggests:
We demonstrated that the algorithm’s performance is comparable to other commonly used tests, such as mammography for breast cancer or cervical cytology for cervical cancer, and its painlessness makes it attractive for repeated testing. A widely accessible smartphone-based tool like this could be used to identify and encourage individuals at higher risk of having prevalent diabetes to seek medical care and obtain a low-cost confirmatory test.
The program is still in the beginning phases, and more trials need to be conducted before it turns into a reliable diabetes-detecting app. Regardless, when it does come out, it could potentially prevent many people from getting diabetes.
