In addition to administering medication, taking vitals, and preparing patients, nurses will overcome their medical history, provide guidance and general support.
Tiffany Kim, assistant clinical professor at Bouve Health Sciences University at Northeastern University, said the “gold standard” method for preparing nursing students for patient interaction is to have patient simulations run standardized patient simulations.
But they have certain logistical restrictions, Kim points out. First, you have to hire an actor. Secondly, you need to find space to do that. And thirdly, professors must report individually after each interaction between each nursing student, which can take time.
To address this, Kim turned to artificial intelligence. Since fall 2024, students in her Public Health Community Nursing course have used AI Simbot, an audio-based simulation program that allows students to conduct mock interviews with virtual patients.
Kim developed AI tools using Openai's large-scale language model and with the help of Northeast alumni Yash Gopaliji Pankhania. It has been updated regularly since then.
The class students were created to help them perform drug use screening following the braft (car, relax, forget, friends, trouble) model.
Students will interact with AI Simbot in two parts. First, they have a conversation with an AI teen named Jordan and ask questions about his alcohol and drug use patterns. Once completed, students will report with an AI assistant named Dr. Casey to discuss how the conversation progressed.
All student conversations with AI Simbot have been recorded and transcribed, allowing students to upload conversations to Canvas, a learning management system for review and grading.
“Students manage their Craff tools while simultaneously practicing their therapeutic communication skills (particularly active listening, empathy, and non-judgmental engagement),” says Kim.
Isabel Iannotty, a recent graduate of the Northeast nursing program, used the tool this summer while enrolling in Kim's class.
She said, “Because we were able to build trust and communicate with each other without affecting actual patients.”
“The chatbot was pretty uneasy at first to answer questions, but this was very realistic,” she says. “Obviously, teenagers who think they are in the doctor's office due to suspected drug use will be a little wary of answering questions honestly.”
Talking to a chatbot cannot be compared to chatting with real people – you cannot read body language or make eye contact with anything eye-opener.
Her conversation with AI Debriefer was also helpful, she says.
“When I interviewed a patient, I flew around with a Craftool [questions] And then you'll ask one question, talk about something, then you'll ask another question,” she says. …I didn't know that and it was actually really useful feedback. ”
Kim says that all professors who teach the courses on Northeastern Boston, Burlington, Fall River and Charlotte campuses have all integrated chatbots into the course.
The chatbot has been experiencing several iterations since Kim released his first beta and is open source. This means that anyone can download and use it from GitHub.
Kim says Simbot allows students to participate in more experiential practices when more loyal simulations are limited when teachers, space and scheduling constraints.
“And it enhances ability-based learning that allows students to exercise more frequently when needed,” she adds, which results in learning, feedback and reflection.
