A new AI-powered chatbot developed by researchers at the University of California San Diego could help patients make more informed decisions about when to seek medical care.
The tool is designed to guide users through self-triage using established medical protocols rather than unverified online information. The approach could reduce unnecessary hospital visits while encouraging patients who need care to seek it sooner, according to an April 23 news release.
The chatbot is powered by conversational AI trained on 100 step-by-step medical flowcharts developed by the American Medical Association. It mirrors how clinicians assess symptoms, asking structured follow-up questions and offering recommendations such as self-care, scheduling an appointment or seeking emergency treatment.
The system relies on three coordinated AI agents working behind the scenes. One identifies the patient’s issue and selects the appropriate medical flowchart, factoring in details such as age and sex. Another interprets patient responses, even when they are not simple yes-or-no answers, to determine the next step. A third translates clinical language into more accessible questions, helping users better understand and respond.
In testing across more than 30,000 simulated patient interactions, the chatbot selected the correct medical flowchart about 84% of the time and followed decision-making steps with more than 99% accuracy, even when symptoms were described in varied ways.
Researchers said it could help reduce the burden on healthcare providers by handling initial triage while allowing clinicians to review conversations and intervene when necessary.
The full study was published April 23 in Nature Health.
The post AI chatbot helps guide patients on when to seek care: Study appeared first on Becker's Hospital Review | Healthcare News & Analysis.
