Principal Investigator: Dr. Sandeep Shirgill (University of Birmingham)
Co-Investigators:
- Dr. Timothy West (Imperial College London)
- Dr. Zhamayne Fakharuzi (E-Mind ltd)
- Mr David P. Cook (E-Mind ltd)
- Mr Daniel Cyrus (University of Surrey)
- Dr. Inocencio Maramba (University of Plymouth)
Lay summary:
The provision of personalised, accurate, and timely information about stroke is a vital component of stroke rehabilitation. Yet research suggests that survivors’ understanding of stroke, its consequences and the support available, remains poor. This is due to problems with how information is currently provided – ranging from complex or irrelevant content, inaccessible physical formats, and static, non-personalized delivery of information. Stroke survivors and carers routinely report dissatisfaction with information provision, and large gaps in availability of relevant and usable information.
We aim to improve the information provision for stroke survivors and their carers using recent developments in large language models (LLMs) to create a chatbot, ASKSiD (Ask for Stroke Information Device) to help patients and carers get accurate, personalised and faster information.
The project will involve co-development of the stroke chatbot with end-users through a series of Patient and Public Involvement and Engagement ( PPIE) activities, as follows:
- Phase 1a: Front-end website and chatbot co-design with end-users
- Phase 1b: Technology development (version 1)
- Phase 2a: Acceptability testing with end-users
- Phase 2b: Product iteration (version 2)
- Phase 3a: Feasibility testing with end-users (version 3)
- Phase 3b: Impact evaluation with end-users
