One of the problems with the COVID-19 virus is that those who are infected can sometimes display symptoms that might be related to other illnesses. For example, coughing is one of the symptoms, along with a fever. This means that at a glance, it would be almost impossible to tell if someone has contracted the virus.However, thanks to the work from a team of researchers from the Mayo Clinic and Nference, it seems that they are turning to the use of AI that will help to isolate phenotypes characteristic of the coronavirus. This means that with the use of AI, it will be able to come up with specific symptoms that show up the earliest in infected patients, meaning that short of actually running a test, these symptoms could be an indicator that the person could be infected.
The general recommendation is that the virus can incubate for about 14 days, but using AI, it can detect early symptoms that could show up 4-7 days before testing. This includes a specific combination of coughing, diarrhea, along with a loss of taste or smell, and excessive sweating.
According to the researchers, “As at-home serology-based tests for COVID-19 with high sensitivity and specificity are approved, capturing these symptoms will become increasingly important in order to facilitate the continued development and refinement of disease models. EHR-integrated digital health tools may help address this need.”
Filed in . Read more about AI (Artificial Intelligence), Coronavirus, Covid-19, Health and Social Hit. Source: venturebeat