Smart Early Diagnosis Symptom Recognition (SEDSR)

When the COVID-19 was announced as a pandemic crisis by WHO, we assigned a task force to explore ideas on how we could support our communities with our disruptive design mindset. Our team of digital developers, biometric data scientists, machine learning and deep learning experts have identified how our existing capabilities could lead to early diagnosis of disease symptoms using regular smart wearable devices.

Looking at the forecast models, the pandemic follows a sine-wave pattern. With the next waves on the horizon, the testing protocols should not only cover for new cases but should also cover for the recovered cases as well. Self-monitoring procedures could narrow down the number of potential cases in need of further testing to use the available resources efficiently.

Most smartwatches, sports wristbands and many cell phones currently are equipped with PPG that is mainly used to measure heart rate and heart rate variability and can readily be used to measure blood oxygen levels. Here, we identified the opportunity to use machine learning and deep learning techniques to define the pattern of changes in blood oxygen levels and cardiac parameters due to the specific effects of COVID-19 on the respiratory and cardiovascular systems in the early stages of COVID-19 infection for the purpose of an early diagnosis.