CeCaS
Supercomputing Platform for Highly Automated Vehicles
PfleDaKi
Data management repository for care-supporting AI applications
Start: 03/2022
End: 02/2025
Care facilities generate large amounts of data – for example, from medical technology, sensor technology, or documentation. This data is rarely considered collectively, although it holds enormous potential. For example, new methods can be developed to care for patients and relieve the burden on nursing staff.
To this end, a platform is being developed in the PfleDaKi project to link data from various care-related sources, such as medical devices, care documentation systems, and electronic devices like smartwatches. The aim is to provide developers of AI applications with accessible data for research and development purposes. The focus is on technical issues such as the form of data storage – centralized vs. decentralized – and the harmonization of heterogeneous data. Nursing science and ethical aspects are also examined. These include the impact of support options on everyday care and the interaction between caregivers and those in need of care. Common standards and well-documented interfaces facilitate the simple connection of different data sources.
The FZI is the project’s consortium leader. It also plays a leading role in developing anonymization and anomaly detection systems in the data streams. Proven techniques such as k-Anonymity based on data structures with industry standards such as HL7-FHIR are used for anonymization. Anomaly detection is important so that the data sets are as easy as possible to use for future AI developers: anomalies are annotated as metadata and provided with the actual data set so that model development can be as robust as possible. Statistical analyses are used for the actual anomaly detection, as well as deep learning-based methods.
Contact
Marc Schroth
Vice Department Manager
Division: Embedded Systems and Sensors Engineering
- +49 721 9654-754
- schroth@fzi.de
- Headquarters Karlsruhe
Research focus
Applied Artificial Intelligence
In this research focus, the FZI prioritizes the topics of Artificial Intelligence (AI) as well as human and AI engineering. In addition, the FZI deals with questions on dedicated AI hardware and predictive AI.
The PfleDaKi joint project is funded on the basis of a decision by the Federal Ministry of Education and Research.
Project partners:
More projects
Transfer Centre for Cybersecurity in SMEs
Cybersecurity for SMEs
PfleDaKi
Data management repository for care-supporting AI applications
InnoSecBW
Innovation with Cybersecurity for SMEs in Baden-Württemberg
FreeSBee
Hardening of embedded RISC-V software by means of code transformations
RackKI
Automatic Integration of Avionik Racks via Artificial Intelligence
WeForming
Buildings as efficient and interoperable components of the future energy system
Hybrenergy
Hybrid building twins for increased energy efficiency
FlexBlue
Flexible refrigeration systems against the background of increased decarbonization
EDIH-AICS
European Digital Innovation Hub applied Artificial Intelligence and Cybersecurity