A&E diagnosis using AI could diagnose patients and cut waiting times

The AI-tool ChatGPT could be used to diagnose patients and reduce waiting times in emergency departments, scientists have suggested.

A study found that the language model, powered by artificial intelligence, “performed well” in generating a list of patient diagnoses and selecting the most likely health issue.

As a test, researchers entered the records of 30 patients who visited an emergency department in 2022, as well as anonymous doctors’ notes, into ChatGPT versions 3.5 and 4.0.

The AI’s analysis was compared with the work of two clinicians, who made an assessment based on the same information, both with and without laboratory data.

When such data was included, doctors had the correct answer in their top five potential diagnoses in 87 percent of cases. The figure was 97 percent for Chat GPT 3.5. The 4.0 version got 87 percent.

Dr Hidde ten Berg from Jeroen Bosch Hospital in the Netherlands said: “We found a lot of overlap with the doctors’ lists of likely diagnoses.

“This indicates ChatGPT was able to suggest medical diagnoses like a human doctor would. We included a case of a patient presenting with joint pain that was alleviated with painkillers, but redness, joint pain and swelling always recurred. In the previous days, the patient had a fever, sore throat and discolouration of the fingertips.

“Based on the physical exam and additional tests, the doctors thought the most likely diagnosis was probably rheumatic fever – but ChatGPT was correct with its most likely diagnosis of vasculitis.”

The team admitted more work is needed before ChatGPT is put to work in a diagnostic setting, as it “is not a medical device” and there are “concerns over privacy” when putting medical data into its system.

But Dr ten Berg added: “There is potential here for saving time and reducing waiting times in the emergency department. The benefit of using AI could be in supporting doctors with less experience, or helping to spot rare diseases.”

The findings, published in Annals of Emergency Medicine, will be presented at the European Emergency Medicine Congress in Barcelona.