Yan Leyfman, Medical Correspondent at OncLive, Freelance, shared on LinkedIn:
“What if an AI conversation could help prepare a physician before the patient even walks into the exam room?
Most discussions about medical AI focus on answering clinical questions, summarizing records, or supporting physician decision-making. But what if AI could also help gather a patient’s history before the visit begins?
A prospective feasibility study evaluated AMIE (Articulate Medical Intelligence Explorer), a patient-facing conversational AI system used before urgent primary care appointments.
Of 114 patients enrolled, 98 completed both the AI interaction and their physician visit.
The findings offer an early glimpse of how this workflow might fit into clinical practice:
- Zero safety stops were required under the study’s predefined criteria.
- Clinical evaluators rated AMIE’s conversations favorably across 87–100% of cases on 17 quality criteria.
- Physicians found AMIE helpful for visit preparation in 33 of 44 cases where they reviewed the transcript before the appointment.
- In 25 of those 44 cases, physicians reported that the interaction might have changed their clinical behavior.
But the study also offers an important reminder: zero safety stops does not mean zero errors. Supervisors identified one hallucination and added clinical information in five interactions.
The bigger opportunity may not be replacing the physician-patient conversation, but making that conversation more productive.
If AI can help patients organize their concerns, communicate their symptoms, and provide clinicians with a useful starting point, it could shift more of the visit toward interpretation, shared decision-making, and care.
The limitations matter: this was a small, single-center feasibility study involving English-speaking adults with a single urgent-care complaint.
Larger and more diverse studies are needed to determine whether these workflows improve diagnostic accuracy, efficiency, safety, or patient outcomes.
The most useful clinical AI may not be the system that makes the decision – but the one that helps the physician make a better-informed decision.”
Title: Conversational diagnostic artificial intelligence in ambulatory primary care: a prospective feasibility study
Authors: Peter G Brodeur, Jacob M Koshy, Anil Palepu, Khaled Saab, Ava Homiar, Roma Ruparel, Charles Wu, Ryutaro Tanno, Joseph Xu, Amy Wang, David Stutz, Wei-Hung Weng, Hannah M Ferrera, David Barrett, Lindsey Crowley, Jihyeon Lee, Spencer E Rittner, Ellery Wulczyn, Selena K Zhang, Elahe Vedadi, Christine G Kohn, Kavita Kulkarni, Vinay Kadiyala, S Sara Mahdavi, Wendy Du, Jessica M Williams, David Feinbloom, Renee Wong, Tao Tu, Petar Sirkovic, Alessio Orlandi, Christopher Semturs, Yun Liu, Juraj Gottweis, Dale R Webster, Joëlle Barral, Katherine Chou, Pushmeet Kohli, Avinatan Hassidim, Yossi Matias, James Manyika, Rob Fields, Jonathan X Li, Marc L Cohen, Vivek Natarajan, Mike Schaekermann, Alan Karthikesalingam, Adam Rodman
Other articles about AI on OncoDaily.