Researchers are reporting promising results from a new trial, which investigated if diabetes could be detected using the photoplethysmography (PPG) signal from smartphones and wearable devices using deep learning.
The study included 22,298 individuals enrolled in the Health eHeart Study, who used the Azumio smartphone application (app), one of the most widely downloaded and used apps to measure heart rate. The researchers developed and applied a deep learning algorithm that used the smartphone-based PPG signal recordings from participants to identify which patients had diabetes based on this signal alone.
They found that the model correctly identified people with diabetes in more than 72 per cent of cases using the PPG signal alone. The test had a strong negative predictive value of 97 per cent. When combining the diabetes score with other commonly accessible risk factors for diabetes, the ability to appropriately classify someone as having diabetes increased to 81 per cent.
Furthermore, after coupling the app-based screening with common risk factors, researchers determined this tool as comparable to many traditional diabetes risk scores that are currently used in clinical practice.
The research will be presented at the American College of Cardiology’s 68th Annual Scientific Session on Sunday.