The tool uses Artificial Intelligence (AI) capabilities including predictive analytics to identify expected clinical coronavirus severity. Researchers hope that the tool will be useful in helping physicians and managing hospital resources when choosing who needs more attention.
An Artificial Intelligence (AI) tool accurately predicted which Covid-19 patients would develop severe respiratory disease. The work was led by New York University (NYU) researchers who successfully revealed the best indicators of future severity.
They also found that these were not that obvious, more work is needed to further validate the model. This search could also lead to another tool that can predict which patients are most vulnerable to the virus.
The tool uses AI capabilities including predictive analytics to identify expected clinical coronavirus severity. Researchers hope that the tool will be useful in helping physicians and managing hospital resources when choosing who needs more attention.
For the study, demographic, laboratory, and radiological findings were collected from 53 patients who have tested positive in two Chinese hospitals. They had mild symptoms such as cough fever and stomach upset. A minority have developed severe symptoms, including pneumonia within a week.
The researchers were surprised when they found that characteristics considered to be hallmarks of Covid-19, such as patterns seen in lung images, fever, and strong immune responses were not useful in predicting severe lung disease, age and gender were not helpful either.
Instead, the new AI tool found that changes in three features, were most accurately predictive of severe lung disease. These were levels of the liver enzymes alanine aminotransferase (ALT) myalgia, and hemoglobin levels.
The team was able to predict severe respiratory disease with up to 80 per cent accuracy. Limitations of the study included the relatively small data sample and the level of severity in studied patients.
A South Korean company used AI to speed up the development of testing kids. The company said that the process would have taken 2 to 3 months if AI technology was not use.
China has also used AI to speed up its scientific work. Scientist decoded the genome sequence of Covid-19 in a month, using AI to speed up the process.