A UK startup says its AI can pass England’s GP exam and go head-to-head with top doctors in a simulated diagnostic environment in what is being heralded as a step forward for humanity.
In groundbreaking research published this week, the Babylon Health technology scored 81 per cent in the GP test against the average passing score for medical students of 72 per cent. It also performed well against clinicians in identifying symptoms and triage in a simulated setting.
The implications, according to Babylon Health founder, could be considerable for access to healthcare worldwide.
“Our mission is to put accessible and affordable health services into the hands of every person on Earth. These landmark results take humanity a significant step closer to achieving a world where no one is denied safe and accurate health advice,” Dr Ali Parsa said in a statement.
The World Health Organisation estimates that there is a shortage of over 5 million doctors globally, leaving more than half of the world's population without access to even the most basic healthcare services.
“Even in the richest nations, primary care is becoming increasingly unaffordable and inconvenient, often with waiting times that make it not readily accessible. Babylon's latest artificial intelligence capabilities show that it is possible for anyone, irrespective of their geography, wealth, or circumstances to have free access to health advice that is on par with top-rated practicing clinicians,” Parsa said.
Working with the Royal College of Physicians in the UK and Stanford Primary Care and Yale New Haven Health in the United States, the company subjected the AI to a recreation of the Member of the Royal College of General Practitioners exam based on publicly available questions, since the test itself is not published.
Babylon Health, one of a number of companies working on triage chatbots for health, has already made a name for itself with a high-profile partnership with the UK’s National Health Service, as well as with an investment round of more than $200 million.
In this latest breakthrough, recognising that real practice differs significantly from a written test, the company subjected the AI and seven primary care physicians to 100 independently-devised symptom sets. Babylon achieved an 80 per cent accuracy on this test while the doctors scored between 64 and 94 per cent. The AI gave a safe triage answer 97 per cent of the time, compared to 93 per cent from the doctors.
The company published the research on its website and hosted a live-streamed event this week to announce the results.
In a recorded demo shown during the livestream, Parsa illustrated how Babylon’s system starts with a chatbot conversation and continues into an AI-assisted telemedicine visit. In addition to informing the doctor’s conversation with the patient, Babylon also automatically generates visit notes for the doctor. It also analyses the patient's facial expressions and reports their emotional state to the doctor.
The company is also working on an Alexa app slated for release at the end of August.
"Tonight's results clearly illustrate how AI-augmented health services can reduce the burden on healthcare systems around the world,” Parsa said.
A version of this story was published on Mobihealthnews, a sister publication of Healthcare IT News Australia.
To share tips, news or announcements, contact the HITNA editor on firstname.lastname@example.org