Several previous studies have proven that smartphones can be used to measure blood oxygen levels (SpO2) in users. Moreover, in 2020, several apps claimed to serve the same purpose and were even available on the Play Store. These apps were later removed due to concerns about their accuracy. Now, a new study by researchers at the University of Washington and the University of California San Diego authenticated these measurements over a wide range of clinically relevant SpO2 levels.
During this study, the researchers were able to accurately measure the levels of blood oxygen saturation up to 70% using a smartphoneCamera and flash module. This proves that users will soon be able to correctly measure their blood oxygen saturation levels with more easily accessible smartphones.
How was this study conducted
Researchers asked study participants to place their fingers on the camera and flash setup of a Google Nexus 6P smartphone. The device’s flash would light up with each heartbeat as fresh blood flowed into that part of the finger. Additionally, the handset’s camera recorded video to measure how much light from the glass was absorbed by the blood across three channels: red, green, and blue.
Before validating the model on other participants, data from four previous participants was used to train a deep learning (DL) algorithm that helped measure blood oxygen saturation. This technology was continued to accurately predict the SpO2 levels of the six subjects whose blood oxygen levels were reduced by the research team using a controlled mixture of nitrogen and oxygen. .
How this technology can be used in the days to come
These researchers have already filed a patent for this technology. Although the technology has only been tested with one smartphone model, the researchers are confident that it can be used on a wider variety of devices in the days to come.
In addition, the technology must be tested on several smartphones because the researchers modified certain configurations on the smartphone to use it to take the measurements. However, the team hopes to continue research by trying the new algorithm on several other test subjects.