Completed Projects

Project Description:

FAIR Data Spaces aims to create a shared, cloud-based data environment that connects science and industry. By integrating the European Gaia-X data infrastructure with Germany’s National Research Data Infrastructure (NFDI), the project establishes a secure, federated framework f​​​​or data exchange and use in line with the FAIR principles—findable, accessible, interoperable, and reusable.

Launched in 2021 and funded by t​he German Federal Ministry of Education and Research (BMBF), the project is driven by a broad consortium of research institutions, universities, and industry partners. It also supports Europe’s broader data strategy through close collaboration with Gaia-X and the EOSC Association.

Key objectives include developing a joint roadmap for Gaia-X and NFDI, clarifying ethical and legal requir​​ements for data sharing, and establishing a unified technical foundation. Several demonstrators illustrate practical applications:

FAIR Data Spaces demo​​nstrates how a secure, interoperable data environment can strengthen both scientific innovation and industrial applications.

Team membe​rs: 

​​Results:

Publications within the FAIR DS Projects can be found on the FAIR Data Spaces Community of Zenodo​ here.​

Project Descripti​on:

Rare diseases (RD) are often difficult to diagnose due to their low prevalence and complex symptoms. Millions of people in Germany and Europe are affected. Delayed or incorrect diagnoses can significantly im​​pair quality of life and delay access to appropriate therapies. One example of such a disease is leukodystrophy, a group of genetically determined metabolic disorders.

Leuko-Expert aims to develop an AI-supp​​orted expert system that assists physicians in the early detection of leukodystrophies. The system is based on a model created using modern data science methods, in particular machine learning.

A proof of concept will be used to verify whether the developed model can reliably support diagnoses. The hypothesis is that an integrative AI model can derive patterns from multimodal data that facilitate earlier and more accurate diagnoses. Decentralized analysis via the Personal Health Train infrastructure ensures data protection. It is intended that the expe​​​rt system will eventually be used across clinics to refer patients to specialist centers more quickly and increase diagnostic certainty. There is also potential to transfer the methodology to other rare diseases.

Team members:

Project Partners:

Res​ults:

A registry has been established at the Aachen, Leipzig, and Tübingen project sites to systematically collect clinical data, genetic findings, and MRI images from patients with leukodystrophies. More than 900 data sets have been entered into the registry to date. Analysis is performed decentral v​ia the Personal Health Train infrastructure, ensuring data protection and efficiency.

Based on this data, AI models were developed that achieve a high accuracy, particularly for symptom data and genetic findings. Due to high heterogeneity, there is still potential for optimization in the MRI data. Initial application experience has shown that treating physicians can obtain diagnoses more quickly and experts can be involved in treatment across locations. In addition, the system opens up the possibility of transferring ​​the developed methods to other rare diseases.

Another result of the project is the Leuko-Expert-Advisor, a digital application that refers patients to centers of expertise for leukodystrophy so that patients with suspected leukodystrophy can be​​ treated more quickly. The Leuko-Expert-Advisor was tested in a real-world environment from August 2024 to July 2025 in a follow-up project called PILEA​.

Proj​ect Description:​​

​​The SMITH consortium (Smart Medical Technology for Healthcare) aims to sustainably enhance patient care by improving the use of clinical data. To achieve this, healthcare data generated at various university hospitals are systematically processed, harmonized, and made available for defined research and analysis projects.

At the core of these efforts are the Data Integration Centers, established as new organizational units at the university medical sites. They serve as the central technological interface: collecting and structuring routine clinical data, converting them into standardized formats, and enabling privacy-compliant, cross-site use in medical research. At the same time, they coordinate agreements among the participating university hospitals to establish common standards for data formats, descriptive systems, and access ​​regulations. All data usage is based on the informed consent of patients, who make a significant contribution to advancing medical care by allowing their data to be used.

Within the SMITH consortium, a network of academic and university medical partners works to purposefully link researc​​h and healthcare. Clinical and methodological use cases are employed to test and demonstrate the added value of the IT solutions developed.

SMITH is one of four consortia funded by the Federal Ministry for Research, Technology and Aerospace (BMFTR) within the Medical Info​​rmatics Initiative (MII). During the expansion and extension phase from 2023 to 2026, the medical informatics infrastructure will be further developed, and collaboration with new partners—particularly in regional healthcare—will be strengthened. This expansion takes place in close cooperation with the Network University Medicine (NUM) to enable even broader data-driven health research in Germany.

Team member​:

​Resu​​lts:

Project Descrip​tion:

The Leipzig Health Atlas (LHA) is a project funded by the German Federal Ministry of Education and Research (BMBF) that pro​​​vides an ontology-based online platform for publications, study data, as well as models and methods. Its aim is to offer researchers, clinicians, statisticians, and other interested parties a quality-assured data atlas to support research and reproducible analyses.

The LHA collects study data and publications, which are accessible upon request, and provides direct implementations for many models to facilitate practical use. The platform builds upon extensive data, methods, a​​​nd expertise from clinical and epidemiological studies, systems medicine research networks, bioinformatics projects, and ontological research initiatives led by partners in Leipzig.

Key objectives of the project include semantic dat​​​a integration, the development of ontologies and analytical services, the validation of source projects, and the integration of all content into a data-sharing platform. Particular focus is placed on the technical CMS platform, the metadata ontology, and the preparation and integration of exemplary project content into the LHA research database.

Team membe​r: 

Results and Publications​​​:

Project Descript​​ion:

PAREMIS was a BMBF funded project in which we designed a concept for model-based registries for clinical research. In particular, we developed a concept for rare diseases including Prader-Willi and Angelman Syndrome and others. PAREMIS was a cooperation between the Leipzig University (IMISE, LIFE), the University medical Center Leipzig (UZSE, IT-Management) and the Institute for Digital Information Technologies (IfDT).

Team member: