Currently, scientific communities and researchers worldwide are striving to develop a line of treatment for coronavirus of COVID-19. And, technology is playing a significant role in such pursuits. The use of artificial-intelligence to speed up the development of line of treatment and vaccines for COVID-19 is an example.
Powered by artificial intelligence (AI), the tool helps to prioritize resources for the promising studies. Also, it helps to ignore research that is unlikely to bear results.
Meanwhile, the U.S. Department of Health and Human Services and the Food and Drug Administration have announced plans to speed up clinical trials, and scientific communities in large numbers are exploring possible vaccines and treatments.
However, the question, which research has the potential to produce much-needed solutions remains.
Nonetheless, for decades, the scientific community has been predicting answer to such questions using Systematizing Confidence in Open Research and Evidence program of the Defense Advanced Research Projects Agency.
Accuracy of Machine Model close to Human Review Process
For endorsement, the program relies on scientific experts to review and rate submitted studies based on their likelihood to be replicable. The average duration for such processes is 314 days – a long period of wait in the midst of the pandemic.
“Regardless, the machine model is as accurate as the human scoring system to make such predictions. And, it can be scaled up to review and rate a large number of papers in a very less amount of time,” say researchers.
“On the other hand, the standard process is very costly, in terms of both financials and opportunity costs,” said the researcher at Northwestern who led the study. Using the standard process, the time involved to move to the second phase of testing is long, and second, it engages experts preventing them from doing their own research.