United KingdomEducation2024 – ongoingCoventry University
Generative AI virtual-patient development with PCS Spark
Coventry researchers produced a virtual-patient draft in a 30-minute activity using generative AI
Problem
Creating virtual patients required extensive scripted questions and answers. The resulting conversations were difficult to adapt.
Approach
Coventry worked with PCS on Spark virtual patients driven by a generative language model. Students and clinicians tested the software and suggested improvements. Researcher Juliana Samson reported producing a better draft in a short activity than with the previous approach.
Results
Takeaway
The reported improvement concerned creating a patient draft. It did not establish better learning outcomes for students.
Categories
- Industry
- Education
- Business function
- IT and R&DProduction and quality
- Innovation type
- Process innovationProduct and service innovation
- AI technology
- Generative AIConversational and voice AI
- Adoption
- Partner co-development
- Evidence
- Company's own claim
Sources
Source summaries
Coventry described its collaboration with PCS and testing with healthcare students and professionals. Samson compared a 30-minute drafting activity with around 100 hours previously needed to create a virtual patient.
Full details are in the original sources.
