AI

Comparing the FeverIQ COVID Risk Classifier

COVID risk estimation is inherently difficult due to the biology of the SARS-CoV-2 virus. FeverIQ COVID Health Check is not only a fast and confidential way to screen for COVID, its risk model trained on millions of symptom reports performs as much as 2.7x better than others, helping businesses and schools minimize the spread of COVID.

The FeverIQ approach to precision COVID risk estimation based on real world data

COVID is dynamic and new insights about its symptoms come every few days. We don’t know when the pandemic will end, but businesses and schools cannot stay closed indefinitely. Although there are a number of risk models and health check systems that seek to screen people for COVID, they tend to have three problems: poor accuracy, lack of privacy, and complicated setup. Enya’s health check system addresses these problems to help businesses and schools reopen safely.

COVID risk estimation is inherently difficult due to the biology of the SARS-CoV-2 virus. The virus has been associated with more than 20 symptoms, which seem to occur in different combinations in different people. More problematically, almost all of those symptoms overlap with those of very common conditions such as allergies, cold, and flu, making COVID hard to reliably diagnose based only on (early/mild) symptoms. The figure shows the performance of 16 different classifiers, including CDC symptom list approaches, a recent algorithm published in Nature Medicine, and the June 5 version of the Enya classifier. Again, the real difficulty with COVID is not needing to detect the likely presence of some kind of respiratory illness, which would be much easier, it's correctly discriminating among four widespread alternatives with broadly overlapping symptoms - cold, flu, COVID and allergies.

Each curve shows how well a classifier can distinguish between people with and without COVID. The straight diagonal line represents a “coin-toss” diagnostic test that correctly classifies people as having COVID 50% of the time. The more accurate the classifier, the further away it will be from the straight line. The green line (Enya’s FeverIQ classifier) is best as judged by its area under the curve (AUC) – it does as much as 2.7 fold better than other classifiers. Enya’s FeverIQ COVID risk classifier is based on data from more than two million people all around the world and gets updated every few days. Based on data availability, the model also considers location and other factors that make it more sophisticated than other risk models.

Beyond a classifier's technical performance, there is another obvious consideration for computing on self-reported data - if a digital risk screening system doesn’t guarantee privacy, people will be reluctant to use it, or if they do use it, they may be reluctant to reveal their true symptoms or give information about other people in their household. For instance, the best predictor of developing COVID in the next few days is if someone in your household recently tested positive. So, we created the first truly private health check system, based on cryptography. We minimize data leakage by only revealing the end result of the risk calculation, not any of the information that went into the calculation (such as your age, location, and specific symptoms you may have). This also reduces a company’s liability as it does not have to handle medical data. If you use the health check system as a guest, the risk estimation is completely private and the final result is known only to you.

By combining a sophisticated risk model with privacy-preserving technology and a simple interface, Enya's FeverIQ effectively estimates a person’s risk of spreading COVID to help the world move forward safely and confidently.

Get updates from our team.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
By signing up you agree to our Privacy Policy.