Virtual Companion
The project is developing an AI-based chatbot based on the social cognitive model and is testing whether it can increase the willingness of cardiac rehabilitation patients to make sustainable lifestyle changes.
Factsheet
-
Schools involved
School of Health Professions
School of Engineering and Computer Science - Institute(s) Nutrition and Dietetics
- Strategic thematic field Thematic field "Humane Digital Transformation"
- Funding organisation BFH
- Duration (planned) 01.01.2026 - 31.12.2026
- Head of project Prof. Dr. Undine Lehmann
-
Project staff
Dr. Karin Haas
Caroline Heuberger
Daniel Reichenpfader
Prof. Dr. Kai-Uwe Schmitt -
Partner
Dr. Thimo Marcin, Berner Reha Zentrum, Insel Gruppe
Prof. Dr. Barbara Solenthaler, Dr. Rafael Wampfler, ETH Zürich - Keywords Rehabilitation, digital care model, behavioural change, chatbot, nutritional intervention
Situation
Digital technologies and artificial intelligence (AI) have the potential to support patients in independently continuing their therapy and implementing a healthy lifestyle after rehabilitation. However, the success of digitally supported aftercare interventions depends largely on the willingness of those affected to change their behaviour and their acceptance of the technology used. Structured, reflective conversations with a chatbot during rehabilitation could strengthen patients' motivation and self-efficacy for a healthy lifestyle change and increase acceptance of digital aftercare services. The project aims to investigate how technology acceptance for the use of a chatbot can be increased and whether the use of a chatbot (virtual companion), which has been further developed on the basis of the social-cognitive Health Action Process Approach model of health behaviour (HAPA model) and motivational interviewing (MI), can increase patients' willingness to make sustainable lifestyle changes during rehabilitation.
Course of action
The project comprises three work packages. In the first work package, a chatbot will be further developed based on the use of a large language model. Aspects of the HAPA model of health behaviour and motivational interviewing will be integrated into the chatbot conversation to guide the dialogue. Depending on the patient's stage of readiness to change behaviour (no intention, intention to change behaviour, implementation), the chatbot will generate individualised and stage-adapted feedback. The focus is on patients with no intention to change. In the second work package, the further developed chatbot will be piloted with 5–10 patients in inpatient rehabilitation. The experiences will be evaluated through surveys (including usability tests, technology acceptance and motivation to change behaviour). In the third work package, a follow-up project proposal will be developed with the active involvement of patients' experiences.
Looking ahead
The project aims to develop a long-term, sustainable and scalable digital intervention for rehabilitation aftercare, thereby closing a gap in care provision. The chatbot is designed to strengthen technology acceptance, motivation for behavioural change and practical implementation during rehabilitation, thereby laying the foundation for long-term self-management. The project thus lies at the intersection of the strategic topics of ‘human digital transformation’ and ‘caring society’ and contributes to the further development of patient-oriented, resource-efficient and sustainable care models.
