Researchers Use AI To Confirm If Someone Has COVID-19 With Just Three Symptoms
A new study (published in journal Computers, Materials and Continua) looked at all the symptoms that were experienced by COVID-19 patients in China to denote the symptoms to specifically look out for, to detect the presence of COVID-19.
We know how notorious COVID-19 really is. Not only is it easy to spread, but people only realise that they¡¯re infected until it¡¯s too late.
Moreover, the symptoms that are normally seen in a COVID-19 patient are similar to someone having a common cold or flu, making it even more difficult to denote the infection in time.
Reuters
Today, this has resulted in over 634 813 active cases across the globe, and despite best attempts, it has managed to take around 30,000 lives. Doctors around the world are striving their best to unravel the mystery of COVID-19 to get a better grasp on the novel coronavirus.
However, a new study (published in journal Computers, Materials and Continua) looked at all the symptoms that were experienced by COVID-19 patients in China to denote the symptoms to specifically look out for, to detect the presence of COVID-19.
Researchers from the US and China used the power of AI to analyse the symptom data from 53 coronavirus patients across two hospitals in Wenzhou in China.
The algorithms were able to denture three changes in the body that highlight severe illness -- body aches, levels of enzyme alanine aminotransferase (also known as ALT) and haemoglobin levels. In case you didn¡¯t know, ALT is a liver enzyme that helps us denote the health of the liver.
The AI going that these three features were the strongest indicators at predicting COVID-19 case. The algorithm was able to show a success rate of 70 to 80 percent in detecting acute respiratory disease syndrome, also called as ARDS. This is the complication that jams a patient¡¯s lungs during COVID-19 and makes it difficult for them to breathe.
Reuters
AI also looked at other patterns -- lung imaging, strong immune responses -- but they didn¡¯t give an accurate result.
According to co-author Anasse Bari, PhD, a clinical assistant professor in Computer Science at the Courant Institute, ¡°Our goal was to design and deploy a decision-support tool using AI capabilities -- mostly predictive analytics -- to flag future clinical coronavirus severity. 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.¡±
