AI distinguishes COVID-19 from influenza and other seasonal diseases

A new study by researchers from the College of Health and Human Services will help determine which

symptoms are more likely to indicate COVID-19. This is especially true during flu season. This is the first study to take into account the seasonality of diseases.

Farrukh Alemi, author of the study, along withcolleagues created a neural network that predicts the likelihood of a patient having COVID-19, influenza or another respiratory disease. The algorithm makes a conclusion based on symptoms in patients of different ages and genders. The study also found that a specific set of symptoms is more important than individual factors. The new tool will help doctors distribute patients even at the stage of admission to the hospital.

The algorithm was created using symptom analysis, aboutwhich were reported by 774 patients with COVID-19 in China and 273 patients in the United States. The neural network was also trained on 2,885 cases of influenza and 884 cases of influenza-like illnesses in patients from the United States.

According to the authors, the next step is to create an AI-powered web calculator that will work in any environment. This will help doctors make a diagnosis before the visit.

Read more:

After ten years of work, scientists questioned the standard model of physics

There is another “planet” inside the Earth: how it saved nascent life

Scientists understand what happens in a dog's brain when a person talks