ARIADNE
ARtificial Intelligence-based Assessment to Detect iNdividuals with Emerging psychosis risk
Dominic Oliver · University of Oxford · United Kingdom
What they're working on
The Clinical High Risk for Psychosis (CHR-P) state offers a vital chance to step in early and prevent the onset of psychosis. But right now, almost 9 out of 10 people in this phase are not identified in time. The best way to detect CHR-P individuals is with a specialist clinical interview. However, these interviews are long, require highly trained clinicians, and are only done in specialist services. Many people who are referred for these assessments do not actually meet CHR-P criteria, meaning a lot of time is spent on unnecessary interviews. Reports are also often difficult for patients to understand and are not always useful for research. To address these problems, we are developing ARIADNE - a new voice-based interview system powered by generative artificial intelligence. ARIADNE will be able to hold a natural conversation, ask the right questions, and produce clear, tailored reports for patients, clinicians, and researchers. This would make assessments faster, more accessible and more empowering for patients while supporting research and clinical care. In this project, we will demonstrate that it is secure, accurate, safe, ethical, acceptable and cost-effective. This will support medical device registration to facilitate ARIADNE helping young people access preventive care.
AI approach
Generative AI · conversational AI · voice AI
Mental health focus
Emerging psychosis risk · early detection