COVID-19 PRESCREENING SYSTEM & METHOD
Categories: “Computer Science“
Reference #: 2020-034
Researchers at Georgetown University have developed a system that can be used as a quick and effective COVID‐19 prescreening tool. The method is based on its ability to determine whether a user is likely to have contracted a disease based on sensor data (compared to a predetermined either population baseline or individualized baseline) received from a user device, such as a smartphone.
More uniquely though, because direct measurement of symptoms using the sensors may not always be feasible or sufficiently accurate, the system also uses surrogates (i.e., indirect evidence that the user is experiencing one of
the symptoms) to identify certain symptoms. For example, the system can identify a fever based on heart rate variability data. It can identify a cough by recording sound and performing sound analysis. It can identify fatigue by analyzing the movement of the smartphone or activity tracker. The system
can also identify shortness of breath (dyspnea) by analyzing respiratory rate data or pulse oxygenation data or recording sound and performing sound analysis. Furthermore, it can identify loss of smell or taste by recording sound and using speech detection algorithms to identify phrases in the recorded sound indicative of loss of smell or taste.
By relying on well established, medically documented, dominant symptoms (and discretizing measures that calibrate them), a machine learnable, interpretable assessment score can be computed, to serve as a comprehensive prescreening tool which identifies whether a user is likely to have contracted a disease.
The weighted function computation is based on symptom prevalence with thresholds determining value meaning: normal, severity (low, medium, high). The determination of the threshold can be based on generalized population norms but can also account for individualized traits. Once a rating, based on
the computed score and assigned thresholds, is determined, triggering actions are activated as appropriate.
Furthermore, as additional research regarding COVID‐19 is published, the system provides a platform that can be
updated so that the thresholds reflect the latest understanding of COVID‐19 symptoms. In the event of a future epidemic or pandemic, the system provides a platform that can also be used to recognize the symptoms of a future virus, bacteria, parasite, engineered bioweapon, etc.
US Patent Application No. 16/878,433
– Notice of Allowance received
Ophir Frieder, Ph.D.
Howard Federoff, M.D., Ph. D.