AI predicts 96% of COVID-19 patients in need of treatment

Corona 19 symptoms vary from patient to patient depending on the severity of the infection. Since the number of beds and intensive care units that can handle patients under the situation of the explosive increase in the number of infected patients is limited, it is important to overcome the spread of infection by taking appropriate measures at the right time for patients. Among these, AI is being developed that can identify COVID-19 patients who need treatment in the intensive care unit with 96% accuracy.

As the number of cases of using AI to cope with the corona 19 epidemic is increasing, attempts are being made to find new production methods for the treatment of Corona 19 using AI or AI that can detect Corona 19 infection from coughing sounds.

Most of those infected with COVID-19 eventually recover, but some suddenly develop severe pneumonia and develop acute respiratory failure, leading to death. AI is being developed to identify these severe corona19 patients.

By combining the AI chest scan image with non-image data such as demographics, vital signs, and blood test results, it succeeded in predicting severely ill patients in need of treatment in the intensive care unit. AI is in the early stages of development, but in order to verify its accuracy, a test was conducted using data from 295 corona19 patients hospitalized in the US, Iran, and Italy. It is said to have been successful.

Other studies have found that ventilator use and potassium levels are the most important non-imaging variables, and the lymphocyte ratio, total bilirubin, and albumin levels are also found to be important variables. In another study, it was found that only non-image data of age, fever, and respiratory abnormalities could predict whether COVID-19 patients should be treated in an intensive care unit.

The AI developed this time focuses on the most important functions in predicting pneumonia caused by COVID-19 after quantifying the range and amount of lung function. Another context factor is used to predict the onset of the most severe symptoms in a patient. The research team hopes to improve AI in the future and use it for testing patients with COVID-19 pneumonia that require closer monitoring and management. In addition, this study did not conduct a test comparing the difference between corona 19-derived pneumonia and respiratory disease. Related information can be found here .