Lancet Study Finds Google's AMIE AI Chatbot Safely Assessed Patients Before Emergency Visits
A Lancet study found Google's diagnostic AI chatbot AMIE safely assessed 98 patients before emergency visits, with no conversations interrupted and 90% diagnostic agreement, according to QbitAI.
In the real clinical study at BIDMC outpatient primary care clinics, 98 patients used AMIE before emergency visits. Supervising physicians monitored every interaction in real time. Under prespecified safety criteria, no conversation needed to be interrupted.
Clinicians reported that AI-generated summaries helped them prepare for consultations in 75% of scenarios. In more than half of cases, the AI output changed the physicians' diagnostic thinking. The differential diagnoses produced by AMIE matched physicians' final confirmed diagnoses 90% of the time.
The report said larger clinical trials are still needed to evaluate how patient-facing AI performs before widespread deployment. Still, the results indicate that AI has the potential to improve doctor-patient relationships and reduce pressure on medical staff. The report added that as Google pushes the boundaries of artificial intelligence in science and health, the work shows meaningful progress. The full paper can be accessed at https://goo.gle/4i7Fx9N.