Smartphones could help detect type 2 diabetesTuesday, 18 August 2020
Researchers at UC San Francisco have developed a “digital biomarker” that would use a smartphone’s built-in camera to detect type 2 diabetes.
The development has the potential to provide a low cost alternative to blood tests and clinic-based screening tools and represents a step forward in diagnosing one of the world’s top causes of disease and death.
Type 2 diabetes affects more than 450 million people worldwide, and can increase the risk of serious health complications like heart disease, kidney failure, blindness and stroke. In the current pandemic, it can also increase the risk of severe symptoms for COVID-19.
Yet, hundreds of thousands of people with type 2 diabetes are unaware they have it and risks to their health.
The vision for the research
“The ability to detect a condition like type 2 diabetes that has so many severe health consequences using a painless, smartphone-based test raises so many possibilities,” said co-senior author Geoffrey H. Tison, assistant professor in cardiology.
“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.”
Screening tools that can use the technology already contained in smartphones, could rapidly increase the ability to detect type 2 diabetes, the researchers said, including populations out of reach of traditional medical care.
According to the World Health Organisation, diabetes is the seventh highest global cause of death, but diabetes can be asymptomatic for a long period of time, making it much harder to diagnose.
Robert Avram, lead author of the study and clinical instructor in cardiology said, “To date, non-invasive and widely-scalable tools to detect diabetes have been lacking, motivating us to develop this algorithm.”
Developing the biomarker
In developing the biomarker, the researchers hypothesized that a smartphone camera could be used to detect vascular damage due to diabetes by measuring signals called photoplethysmography (PPG), which most mobile devices, including smartwatches and fitness trackers, are capable of acquiring.
The researchers used the phone flashlight and camera to measure PPGs by capturing colour changes in the fingertip corresponding with each heartbeat.
In the Nature Medicine study, UCSF researchers obtained nearly 3 million PPG recordings from 53,870 patients in the Health eHeart Study who used the Azumio Instant Heart Rate app on the iPhone and reported having been diagnosed with diabetes by a health care provider. This data was used to both develop and validate a deep-learning algorithm to detect the presence of diabetes using smartphone-measured PPG signals.
Overall, the algorithm correctly identified the presence of type 2 diabetes in up to 81 percent of patients in two separate datasets. When the algorithm was tested in an additional dataset of patients enrolled from in-person clinics, it correctly identified 82 percent of patients with type 2 diabetes.
Among the patients that the algorithm predicted did not have type 2 diabetes, 92 to 97 percent indeed did not have the condition across the validation datasets. When this PPG-derived prediction was combined with other easily obtainable patient information, such as age, gender, body mass index and race/ethnicity, predictive performance improved further.
Diagnosing diabetes in the future
At this level of predictive performance, the authors said the algorithm could serve a similar role to other widespread disease screening tools to reach a much broader group of people, followed by a physician’s confirmation of the diabetes diagnosis and a treatment plan.
“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,” said study author Jeffrey Olgin, UCSF Health cardiologist and professor and chief of the UCSF Division of Cardiology.
“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 authors recommend further study to determine the effectiveness of this approach for specific clinical applications, such as screening or therapeutic monitoring.