A pilot study presented at the European Emergency Medicine Congress reveals that ChatGPT, an AI chatbot, performed comparably to a trained doctor in suggesting likely diagnoses for patients in emergency medicine departments. Although further research is needed, these findings suggest that AI technology like ChatGPT could potentially assist doctors in emergency medicine, potentially leading to shorter patient waiting times. The study, conducted at Jeroen Bosch Hospital in the Netherlands, involved anonymized data from 30 patients. ChatGPT generated likely diagnoses that overlapped with doctors diagnoses in around 60% of cases. Doctors had the correct diagnosis in their top five likely diagnoses in 87% of cases, compared to 97% for ChatGPT version 3.5 and 87% for version 4.0. While the technology is not a medical device and raises privacy concerns, it holds potential for supporting less-experienced doctors and identifying rare diseases, ultimately improving patient care in emergency departments.
Global News and Digital Insights
for the Healthcare Industry
Menu
