BehAIve

Behaviour Monitoring and Support of Older Adults 

_{Subproject: User-Centered Development of Assistance Systems and Explainable Voice Interfaces}

Germany is undergoing substantial demographic change, characterised by an ageing population and increasing life expectancy. As a result, the prevalence of chronic diseases and dementia, as well as the demand for long-term care, is increasing considerably, while the shortage of professional caregivers continues to grow. This development is particularly pronounced in Mecklenburg-Western Pomerania, where the proportion of older adults is increasing at an above-average rate. Innovative digital and AI-based assistive systems may help older adults maintain their independence and health for longer. The early involvement of affected individuals, family caregivers, and healthcare professionals is essential for the successful development of such technologies.

The aim of BehAIve is to develop an AI-based assistive system that can provide context-adaptive support to older adults during hospital stays and to people with cognitive impairment in their home environment. Through continuous monitoring and the detection of potentially hazardous situations, the BehAIve system is intended to support independent living at home, delay the need for transition to residential care, and help maintain autonomy and social participation. In this way, the system could have the potential to improve the quality of life of people with cognitive impairment and geriatric patients while simultaneously reducing the burden on professional and family caregivers.

To achieve these objectives, BehAIve will combine AI methods for time-series analysis and symbolic modelling with data-driven deep-learning approaches. In addition, system explainability, user acceptance, and ethical and data-protection considerations are investigated and integrated into the development process.

DZNE Rostock/Greifswald Subproject

DZNE Rostock/Greifswald is responsible for the user-centred development of the system for people with mild cognitive impairment (MCI). This includes, in particular, the design and application of participatory technology co-design methods to systematically integrate the needs and requirements of the target population into the development process.

In addition, DZNE Rostock/Greifswald is responsible for developing explainable voice-based user interfaces. The aim is to design transparent and comprehensible interaction concepts that promote user acceptance and usability and enable appropriate support for users.

_{Project Partners 
Prof. Kristina Yordanova, Institut für Data Science, Universität Greifswald (Projektleitung) 
Prof. Thomas Kirste, Institut für Visual & Analytic Computing, Universität Rostock 
Prof. Frank Krüger, Fakultät für Ingenieurwissenschaften, Hochschule Wismar 
Prof. Ralf Schneider, Rechenzentrum, Universität Greifswald 
Prof. Agnes Flöel, Klinik für Neurologie, Universitätsmedizin Greifswald 
Prof. Maximilian König, Klinik und Poliklinik für Innere Medizin, Universitätsmedizin Greifswald 
Dr. Joschka Haltaufderheide, Institut für Ethik und Geschichte der Medizin, Universitätsmedizin Greifswald}

_{Associated Partners 
Dr. Emma Tonkin, University of Bristol 
Jun.-Prof. Stefan Lüdtke, Universität Rostock 
Prof. Giovanni Rubeis, Universität Graz}