The University says the use of ‘next-level’ artificial intelligence means students will be able to interact with an avatar that ‘behaves like a patient’ by giving individualised answers to specific questions, rather than just being able to list symptoms. The new large language model AI also ‘drastically cuts the time it takes to create a virtual patient’, meaning eventually people will be able to easily create their own customised situations.
PhD student, Juliana Samson, from the University’s Research Centre for Healthcare and Communities, worked as a physiotherapist for 20 years, and is leading the research alongside AI firm, PCS, to create the software which ‘learns as it goes’, allowing the creation of more sophisticated avatars for students to practice with and learn from.
Juliana said: “Previously you would have to create folders and folders of every single question and answer that you might imagine happens in a conversation, and link it all together – it was really laborious, and wasn’t as fluent or adaptable. Now we’re massively cutting the time this needs in terms of the build and the responsiveness is off the scale.
“With this artificial intelligence we have been able to create an avatar that actually behaves like a patient, and can be created quickly. It would previously have taken around 100 hours to create a virtual patient, but we did a 30-minute activity and the draft was miles better than before. The software has stepped up a level. The AI can generate a lot of human conversations that you would need to have, and students will be able to have conversations without traumatising or upsetting a real human, and if you get it wrong, you can go again.”
The conversations see students given points for asking pertinent questions on different subjects, as well as being able to read the entire conversation back afterwards to look at what went well and what they could improve.
For her current research, which also involves Coventry University’s Research Centre for Postdigital Cultures, Juliana is working with students and professionals in healthcare and technology. Several healthcare disciplines are involved – spanning from paramedics to social work. Another benefit of the technology is helping students practice taking a patient’s history during a clinical conversation at an earlier stage in their career, as this would not usually happen until on placement.
Balazs Moldovanyi, CEO of PCS, said: “We are thrilled to provide Coventry University with our most intelligent Spark virtual patient AI, now driven by an in-house trained generative large language model for never-before-seen realism in communication training. We are excited to see how far the university’s domain experts can take this AI with their custom scenarios.”
Michelle Castleberry, co-founder of PCS, said: “Collaborating with Juliana is a dynamic experience, and her enthusiasm and mission-driven mindset infuse every meeting with purpose and energy. She adeptly learns the more technical aspects of our technology, while educating our team on the necessary clinical components of the project.
“I feel privileged to be a part of this collaboration, and together we’ve synergised our strengths to achieve ambitious milestones over the past 20 months, while staying true to our shared goal of making a positive impact in Interprofessional Education (IPE) using AI-driven simulations.”
The latest step in improving the virtual patients is a co-creation event to optimise the performance of the AI with a team of occupational therapy and physiotherapy students, academics, and clinicians, using the software to suggest ways in which it can be improved or made more realistic.