Euronews: Artificial intelligence screening patients at risk for pulmonary injury with COVID-19
The team behind the study reported 80% accuracy in their predictions about which patients would develop acute respiratory distress syndrome (ARDS) – which can be fatal in severe cases with COVID-19, Euronews reports.
The study also follows another study published earlier this month. It states that out of 201 patients with pneumonia at a hospital in China, 84 developed ARDS, 67 received mechanical ventilation and 44 died. All of those who died died of ARDS and received mechanical ventilation.
The purpose of the study is to provide hospitals with a tool to help decide which patients can be safely sent home and who will need beds and potential hardware.
The AI tool used data from 53 patients from two hospitals in China, all with positive coronavirus samples in January.
* ARDS is a condition in which the lungs, inflamed by a serious infection such as pneumonia, cannot provide enough oxygen to vital organs in the body.
The condition forms air balloons in the lungs, making breathing difficult. It is the cause of death in many of the fatal cases of coronavirus, with severe cases of pneumonia damaging the patients’ lungs.
A study by scientists at New York University’s School of Medicine and the Institute of Mathematical Sciences, published in the online journal Computers, Materials & Continua, found surprising results – the best indicators of future weight are not expected.
Instead of factors such as certain models in imaging of the lung, fever and even age and gender, changes have been found in three characteristics that give the most accurate prediction for future deterioration.
One is the levels of the liver enzyme alanine aminotransferase (ALT).. Although they increase dramatically when the disease damages the liver, they are only slightly higher in patients with COVID-19, but these slight changes are key to predicting severity.
The second is deep muscle painassociated with greater overall inflammation in the body.
Higher hemoglobin levelscontaining iron, which allows blood cells to transport oxygen to body tissues, is also associated with later respiratory distress.
The small sample size of the study limits its current usefulness, but researchers believe it is “promising as a tool for predicting patients most vulnerable to the virus,” according to one author, Megan Coffey.
“We hope that when fully developed, the tool will be useful for doctors as they will evaluate which moderately ill patients really need beds and which ones can safely go home. Which, in turn, will alleviate the problem of shortage. hospital resources, “said co-author Anas Barry.