AI-Powered Virtual Patients Help Dental Residents Practice the Conversations That Matter
核心洞察
A University of Rochester collaboration is piloting SOPHIE (搜索), an AI-powered virtual patient platform, to strengthen dental residents' communication, trust-building, and empathy skills.
The training, organized around the "three E's" of Empower, Empathy, and Explicit, lets residents practice difficult conversations repeatedly in a low-stakes, browser-accessible setting.
Residents reported increased confidence in handling challenging patient conversations, with AI-generated measures suggesting improvements in conversational balance and open-ended questioning.
A University of Rochester collaboration is using an AI-powered virtual patient to help dentists strengthen communication, build trust, and better respond to patients' fears and concerns. The interactive platform, called SOPHIE (搜索)—Standardized Online Patient for Healthcare Interaction Education—was adapted for dental education through a partnership between the Eastman Institute for Oral Health (搜索) (EIOH) and the University's Department of Computer Science.
"Fear, anxiety, uncertainty about a procedure, concerns or misconceptions about treatment can all shape a patient's experience," said Shasha Cui, EdD, MBA, assistant professor and education specialist at Eastman Institute for Oral Health (搜索). "For dentists, providing excellent care requires more than technical expertise. It also requires knowing how to listen, explain complicated information clearly, recognize concerns and establish trust."
A Scalable Alternative to Standardized Patients
SOPHIE (搜索) uses a three-dimensional virtual character that can see and listen to a clinician, carry on a conversation, and respond in real time. The technology was originally developed to help physicians practice difficult conversations with patients facing serious illness, and Cui saw an opportunity to adapt that same approach to dentistry.
"Health professions programs have traditionally used trained actors, known as standardized patients, to help clinicians develop communication skills," Cui explained. "The approach can be highly effective, but it also requires numerous resources, making it difficult to provide every learner with repeated opportunities to practice."
Because SOPHIE (搜索) can be accessed through a browser, trainees can practice at their convenience and repeat scenarios as often as needed. That repeatability matters when the skills being practiced are inherently difficult: recognizing anxiety, active listening, showing empathy, asking open-ended questions, and involving patients in decisions about their own care.
The Three E's: Empower, Empathy, and Explicit
The team organized the training around what it calls the three E's: Empower, Empathy, and Explicit. The framework encourages clinicians to involve patients in their care, demonstrate empathy, and communicate information clearly. The training allows residents to try different approaches, receive feedback, and practice again.
"AI cannot replace empathy," said Ehsan Hoque, PhD, professor of computer science and an expert in human-centered artificial intelligence, "but it can provide an environment where clinicians can repeatedly practice the communication skills patients depend on—without the pressure of practicing for the first time during a real clinical encounter."
Cultural differences and language can also influence how patients perceive treatment, ask questions, and interact with their providers. "SOPHIE (搜索) really helps us understand how to communicate better with the wide range of patients Eastman serves," said Prateek Daga, DDS, an EIOH resident who participated in the pilot program at EIOH's Golisano Specialty Care Perinatal Dental Clinic.
For Daga, communication is especially important because clinicians may have only a limited amount of time to establish a relationship with a patient. If that connection is not made, he said, trust can quickly be lost.
Measurable Improvements in Resident Confidence
Residents reported increased confidence in handling challenging patient conversations and valued the ability to practice repeatedly in a low-stakes setting. AI-generated measures also suggested improvements in conversational balance and the use of open-ended questions.
The collaboration reflects the University's priority for implementing Team Science—experts from different disciplines applying their knowledge to a problem none could address as effectively alone. "The ability to collaborate across disciplines has been extremely beneficial for many projects and is one of the University of Rochester's strengths," said Jin Xiao, DDS, PhD, EIOH professor and associate dean for Oral Health Education at the University of Rochester School of Medicine and Dentistry.
The study, published in the Journal of Dental Education, brought together researchers from Eastman Institute for Oral Health (搜索), Computer Science, and the Clinical and Translational Science Institute.
"The SOPHIE (搜索) project has been a successful and exciting step forward in exploring how AI-powered virtual patients can enhance communication training for dental residents," Cui added. "The positive response from our residents and the lessons learned from the pilot have demonstrated how this approach provides meaningful, practical, and scalable learning experiences. We look forward to continuing our collaboration with our colleagues to further develop SOPHIE and expand its use for resident education."
For dentists and educators at EIOH, the ultimate purpose of the technology is much simpler than the AI behind it. "Great dentistry is not only about fixing teeth," Xiao said, "but also caring for our patients."