What we do
DIFUTURE Tübingen joined the GO:FAIR PHT Implementation Network in 2018 as a milestone for distributed FAIR analysis. We developed a productive deployment-ready PHT-meDIC solution to achieve the defined goals from the PHT manifesto.
Our architecture and services are fully open source accessible, and demonstrate how modern cloud techniques can be leveraged for complex distributed, privacy-preserving medical data analytics.
Learn more about our current work
We work on several use cases with different partners to evaluate and demonstrate our developed PHT version.
Our current main focus is:
- deployment
- security methods extensions
- image and genome analysis
Jun ’22
A PHT-meDIC presentation followed with an interactive workshop. Presented at the MIRACUM & DIFUTURE Summer School 2022 by Marius de Arruda Botelho Herr and David Hieber (slides in German).
Presentation (google)
Nov ’21
A PHT Tübingen CORD demo presentation presented by Marius de Arruda Botelho Herr. Presented at the 16. CORD-MI-WebWorkshop (slides in german).
Presentation (google)
Dec ’20
A PHT Tübingen update presented by Lukas Zimmermann. Presented at International FAIR Convergence Symposium 2020.
Oct ’20
A shared status update with Aachen presented by Marius Herr, Sascha Welten and Oya Beyan with the architecture from Tübingen, Demo in Aachen and next steps. Presented at the MII workshop for distributed analysis in Germany.
Oct ’19
A status update presented by Marius Herr of the recent progress at Tübingen. Presented at the DIFUTURE symposium 2019 in Tübingen.
Nov ’18
The first proof-of-concept presentation by Oliver Kohlbacher of our PHT architecture, workflow and train-API. Presented at the DIFUTURE symposium 2018 in Munich, the GO FAIR meeting in Leiden and Berlin .
Our code is fully open source and can be accessed at github.com/PHT-Medic.
Documentation available at pht-medic.github.io/documentation.
Our current main for extensions are:
- Deploy distributed learning architecture at projects partners on national level
- Grow a user community
- Extend towards federated learning and SMPC
Our Team
Michael studied in his B.Sc cognitive science and finished his M.Sc in computer science in 2021. He works as research assistant at the Institute of Translational Bioinformatics (TBI) and works as technical lead on: security implementation and enabling new methods using the Personal Health Train.
David studied computer science in Tübinen and finished his M.Sc in 2021. He works as a research assistant at the Institute of Translational Bioinformatics on the PHT and Leuko-Expert projects.
Peter studies media informatics (M.Sc) and works for the hospital (UKT) as IT-Worker and is also responsible for the e-learning and survey applications. At TBI he works as a research assistant. He contributes to the following PHT topics: security implementation, UI- designs & applications, backend APIs and microservices.
Mete (Ph.D. in computer engineering), works as Postdocotoral researcher and security advisor on the Personal Health Train Tübingen. He is part of the DIFUTURE analytics Team.
Collaborate
You are interested to contribute to the PHT project in one of the following ways:
- research questions you want to answer with de-central analysis
- concepts to extend or enable different methods
- code development