UC San Francisco researchers designed a smartphone app that can measure an individual's heart rate through the phone's built-in camera and a photoplethysmography signal, which can detect blood volume changes, to flag individuals who may have diabetes.
In its early stages, diabetes can alter an individual's blood vessels. Researchers used machine learning and a smartphone camera to detect blood volume variations by shining the smartphone flashlight on an individual's fingertip. When the heart contracts, blood pressure increases in the vessels, causing them to expand. This then increases the amount of light reflected by the skin to the optical sensor of the phone's camera.
"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," Robert Avram, MD, lead study author, said in a news release
The study comprised more than 54,000 individuals at an average age of 45 years old. Participants enrolled in UCSF's online Health eHart Study, which uses Azumio's smartphone app designed to instantly measure the user's heart rate. Researchers applied a deep learning algorithm that used the smartphone-based PPG signal recordings from participants to identify those who had diabetes.
Results showed that the model correctly identified participants suffering from diabetes in more than 72 percent of cases using the just the PPG signal.
"The potential to transition screening that's normally done by physicians or nurses to the patient themselves through a smartphone app is a very novel concept and gives us a glimpse into how healthcare might work in the future," Dr. Avram said.
Study authors concluded that they are hopeful about the use of this technology to detect diabetes, but a physician is still required to confirm a diabetes diagnosis.