bim4cAIre – Shaping the Future of Care with the Digital Twin

bim4cAIre Vanessa Liebler, CC BY SA 4.0

The objective of the research is to develop automated processes for the generation of a digital building twin for a specific digital care usecase (e-health). The focus of the research is divided into two main areas - the evaluation of consumer-based technologies for 3D data acquisition on the one hand, and the development of software modules using machine learning methods for analysing the acquired data on the other hand. With the project’s outcome, a beneficial and prototypically implemented concept is expected, which is going to automate the laborious process largely and thus contribute to a socially important issue.


In light of the individual demands of an aging and increasingly single population, the bim4cAIre project team is developing concepts for a digital and sustainable care process. In this context, care at home is a relevant process, which is directly related to the identification of barriers in living spaces. To this end, novel methods for the 3D acquisition of complex living spaces as well as adaptive methods for the analysis of the acquired data are examined. The challenge relating with the usecase care at home is the high degree of individuality - in terms of the living space and the users addressed.


The spectrum of activities is multifaceted. The initial phase of the project aimed at selecting a diverse bunch of current technologies for 3D data acquisition and evaluating them against the requirements of care at home. These results are essential as they provide the foundation for further work. Integrating accurate and available technologies for data acquisition into the identification process of barriers requires further software modules in terms of data enrichment and creating explainable statements about the accessibility of the digitally represented living space. Results of this work have been presented at (inter-) national conferences and published in journals.


Within the bim4cAIre project, the potential of consumer-based technologies for 3D data acquisition could be demonstrated. Based on this, a mobile application called Semantic Data Capture was developed, which is able to acquire and classify 3D data simultaneously. The application represents a novel method in order to deliver intelligent 3D point clouds on-demand. In order to identify barriers and to communicate them to affected people in a low-threshold and transparent way, a bunch of machine learning methods processes the resulting data of this application. Beyond this explained usecase, partial results of the project may also have the potential to contribute to specific issues in neighboring disciplines, like architecture engineering and construction industry. For example, Semantic Data Capture could be used to replace largely manual activities out of the Scan-to-BIM ecosystem.