Motivation:
The AutoCaFe (“AUTOmated CAre Feedback”) project aims to improve the quality of care through an AI-supported, sensor-based monitoring system. In light of the shortage of skilled workers and demographic change, there is a growing need for technical support that relieves the burden on nursing staff and improves care for older people. Technical assistance systems are considered a promising solution, but they raise questions of data protection, acceptance, and ethical acceptability. There is particular concern that standardization and technologization could jeopardize relationship building and the understanding of personality in care.

Goal:
The aim of the project is to develop an AI-supported system for the automated recognition of nursing activities. Based on sensor-based data, the system should be able to distinguish between different activities and indicate where support is needed. The information collected will be used to facilitate communication, planning, and documentation. The focus is on designing technical assistance systems that reflect nursing ethics—in other words, systems that do not restrict degrees of freedom, individual routines, and the social dimension of nursing, but rather make them visible.

Approach:
The approach combines sociological and computer science research in an interdisciplinary co-design. Multisensor systems (e.g., cameras, body tracking sensors, audio sensors) are used, and their data is analyzed using generic and individual classifiers. The accompanying sociological research uses ethnographic methods to reflect on the development process and critically accompany the integration of care values into technical systems.

Funding: 
The project is funded by the German Research Foundation (project number: 536726758).

Partners:
Zeppelin Universität Friedrichshafen, Prof. Dr. Florian Muhle