Researchers from the University of Washington and the University of California San Diego have come up with an unusual way to use a smartphone. It turned out that using this gadget, you can measure the level of blood oxygenation. The accuracy of this method was over 80%.
The scientists used a convolutional neural network to train an AI based on 10,000 levels of oxygen in the blood from 61% to 100%. The user just needs to turn on the flash and the camera, and then put a fingertip on them. The camera records how much blood absorbs flash light in each of the three measured color channels: red, green, and blue. The collected data is used for AI training. It is assumed that over time, the accuracy of measurements will increase.
The researchers tested the model on volunteers. Each had their oxygen levels measured using a traditional pulse oximeter and a new app. According to the results, the smartphone measures in the clinical range of 70% to 100% (like most pulse oximeters), while the accuracy is over 80%.
So far, we are not talking about the medical use of smartphones. Although it can be a great option for developing countries and remote regions.