San Francisco, CA – Over 30 million people in the U.S. are living with diabetes, however, one in four are unaware of their condition. Diabetes progresses over time and when left untreated, it can have serious health consequences such as kidney damage, eye problems and an increased risk for heart disease or stroke. Understanding a consumer need for easier access to healthcare, researchers from the University of California in San Francisco are using data from a recent study to explore a new smartphone app that can provide a diagnosis without a trip to the doctor.
“After coupling the app-based screening with common risk factors for diabetes (age, gender, ethnicity and body mass index) our tool was comparable to many traditional diabetes risk scores that are used in clinics to predict diabetes and does not require a physician to administer,” explained Dr. Robert Avram, the lead author of the new study.
A number of fitness apps already on the market use a smartphone’s camera and flash through technology referred to as a photoplethysmography (PPG) signal, researchers realized that this tech could also be used to observe the minute-by-minute changes resulting from the expansion of blood vessels in a person’s finger. Through this data, it makes it possible to uncover information about a person’s blood flow, which is necessary in diagnosing the condition.
In testing their new discovery, scientists created a deep learning algorithm that used the Azumio Instant Heart Rate smartphone app’s PPG signal to determine which of the study’s participants had diabetes. Coupling the results with other known risk factors and pitting the information against their database, the algorithm was able to correctly identify diabetes 81 percent of the time. On the other hand, the algorithm was also able to correctly identify participants as not having diabetes 97 percent of the time.
“Based on our findings, this strategy could become a low-cost way to screen for diabetes at home because it can be derived from any optical system that has a camera and a flashlight, and most people have a smartphone," stated Dr. Avram.
Following the presentation of their findings on Sunday, March 17, at the American College of Cardiology's 68th Annual Scientific Session in New Orleans, LA, the researchers plan to take their study a step further by testing the algorithm against different populations and use it to determine if someone has early- or late-stage diabetes.
Why This Matters –
Today, more than 40 million Americans cannot afford or access healthcare and as a result, serious health conditions such as diabetes are going undiagnosed and untreated. Statistics like this are driving a demand for alternate diagnostic tools. From your home to the doctor’s office, tech that can provide a diagnosis no matter where you are is fueling the creation of these new alternatives. So far this year we have seen a number of new smartphone apps and tech driven tools come to the market that aim to provide health care options that are not only more affordable, but easier to access.
Have you checked out our 2019 Health Trend Ten? System of One (Trend 9), Diagnosed Anywhere (Dynamic 1) uncovers how increasingly sophisticated diagnostics are going to change conversations about care at home, within retail environments and in primary care.