Artificial Intelligence Used To Predict Which Coronavirus Patients Are At Greater Risk Of ARDS
An artificial intelligence tool is being developed which researchers hope could be used to predict which coronavirus patients will suffer life-threatening lung damage.
The team behind the initial study reported 80% accuracy in its predictions of which patients would develop acute respiratory distress syndrome (ARDS) – which can be fatal in severe COVID-19 cases.
The aim of the research is to provide hospitals with a tool to help decide which patients cans safely be sent home and which will need beds and potentially breathing equipment allocated to them.
The AI tool used data on 53 patients from two hospitals in China, who all tested positive for coronavirus in January.
ARDS is a condition where the lungs, inflamed by serious infection such as pneumonia, cannot provide the body’s vital organs with enough oxygen. The condition causes fluid to to leak into the air sacs in the lungs, making it difficult to breath.
It is the cause of death in many fatal coronavirus cases, with severe cases of pneumonia damaging the lungs of patients. In one study published earlier this month, of 201 patients with pneumonia in a hospital in China, 84 developed ARDS, 67 received mechanical ventilation and 44 died. All those who died developed ARDS and received mechanical ventilation.
The AI study, conducted by researchers at NYU Grossman School of Medicine and the Courant Institute of Mathematical Sciences at New York University, looked at demographic, laboratory and radiological findings of patients with COVID-19.
The paper, published online in the journal Computers, Materials & Continua, found the best indicators of future severity were not as expected.
Instead of factors such as certain patterns in lung imaging, fever, and even age and gender, it found changes in three features gave the most accurate prediction of future deterioration:
- Levels of the liver enzyme alanine aminotransferase (ALT). While these rise dramatically when diseases damage the liver, they were only a bit higher in COVID-19 patients, but these slight variations were key to predicting severity.
- Deep muscle aches, linked to higher general inflammation in the body
- Higher levels of hemoglobin the iron-containing protein that enables blood cells to carry oxygen to bodily tissues, were also linked to later respiratory distress.
The small sample size of the study limits its current utility, but the researchers think it “holds promise as another tool to predict the patients most vulnerable to the virus,” according to one of the authors, Megan Coffee MD.
“We hope that the tool, when fully developed, will be useful to physicians as they assess which moderately ill patients really need beds, and who can safely go home, with hospital resources stretched thin,” said co-author Anasse Bari.
Marr, B. (2019, March 6). The Revolutionary Way Of Using Artificial Intelligence In Hedge Funds. Retrieved from https://www.forbes.com/sites/bernardmarr/2019/02/15/the-revolutionary-way-of-using-artificial-intelligence-in-hedge-funds-the-case-of-aidyia/#3274b36357ca