Back in 2016 we posted about how IBM’s Watson was helping to diagnose previously unknown conditions.
Well, AI is at it again, this time to provide symptom analysis for COVID-19. These systems have actually been around for quite a while, but they’re relatively unknown by those not in the medical field and their underutilisation during the pandemic may be putting an undue burden on healthcare workers.
These systems learn from anonymised medical data sets and patient records, as well as notes from consultant physicians to help triage potential patients.
There have been multiple peer-reviewed studies on these sorts of systems, such as this review from Baker et al (2020) which found that patient assessment and diagnostic information was comparable to that of a human physician.
“Overall we found that the Babylon Triage and Diagnostic System was able to identify the condition modeled by a clinical vignette with accuracy comparable to human doctors… the AI system was able to safely triage patients without reverting to overly pessimistic fallback decisions.”Baker, et al (2020)
The University of California, San Francisco (UCSF) have taken these programs one step further by fully integrating with patients personal medical records, to enable immediate appointment scheduling.
There are many programs out there, most of them freely available, but one study from the University of Waterloo has recently identified that ~50% young adults (the group most likely to lead the uptake for novel technologies) are not aware that these apps exist.
The links above will take you to the websites of Babylon and Isabel so you can find out more for yourselves, and maybe give them a try. Healthcare systems across the globe are experiencing the greatest demand of a generation, and anything the general public can do to limit their uptake of these limited resources is worth considering!