As initially reported by France24, researchers from the United States and China have discovered that artificial intelligence may play a crucial role in understanding the impact of COVID-19 on patients. In particular, AI could allow doctors to recognize which patients with COVID-19 may later develop acute respiratory distress syndrome (ARDS). You can find the full research article in Computers, Materials & Continua.
About Acute Respiratory Distress Syndrome
Acute respiratory distress syndrome (ARDS) is a severe and fairly sudden-onset lung disease. It usually occurs within 1-2 days after injury or illness. ARDS prevents oxygen from getting to the lungs and blood because of a fluid buildup in the alveoli. The disease manifests through symptoms of shortness of breath or breathing issues, fatigue, cyanosis, and organ failure.
Patients with ARDS will also experience stiffness of the lungs, and many individuals with ARDS face organ failure to the kidneys or liver. While ARDS tends to affect older individuals, it can affect any patient with a severe lung illness or injury; in the current medical climate, this relates directly back to COVID-19.
You can find out more about acute respiratory distress syndrome on our website.
Artificial Intelligence
Researchers used an algorithm that could allow doctors to recognize which patients are more at risk. This helps doctors prioritize care. First, they took data from 53 patients with COVID-19 from China and applied the algorithm to this data.
Researchers found that three distinct features were predictors of acute respiratory distress syndrome. First was body aches. Next came hemoglobin levels; elevated red blood cell levels indicated possible disease continuance. Finally, elevated levels of alanine aminotransferase (ALT), a liver enzyme, also suggested that a patient might develop ARDS.
Surprisingly, it was also discovered that factors like age, sex, fever, and immune responses in lung images were not predictors of further disease. Megan Coffee, who worked on the study, stated:
“A lot of the data points that the machine used to help influence its decisions were different than what a clinician would normally look at.”
Ultimately, this AI tool was able to predict a patient’s risk of developing acute respiratory distress syndrome with up to 80% accuracy. While the team is still working on adding data and creating a more accurate and effective tool, the goal is to have it available for doctors to use in April 2020.
