We study the energy metabolism of the human brain and how it relates to neural signaling in the healthy and diseased brain at Technische Universität München (TUM) in Munich, Germany.
This website serves as a guiding tool to familiarize yourself with a way of conducting imaging research and data analysis by means of a coordinated data pipeline.
Why is this important?
Imaging research produces vast amounts of data from different technologies in different formats. Hereby, data is stored and organized differently by every researcher, making the traceability of data difficult. Additionally, data is analyzed heterogeneously using different software or different versions of the same software. Likewise, preprocessing analysis pipelines are also heterogeneous for the same type of data and their results are stored differently for every researcher.
Therefore, we implemented a data management system based on an open source solution which homogenizes the data handling from its collection until it is analyzed using imaging standards:
- Automatizing and centralizing the data management
- Standardizing and harmonizing analysis workflows
How does it work?
The harmonization of the image analysis workflows is based on the embedding of different imaging software into containers that can be run automatically and directly from the server. Finally, the aggregation of the individual analysis results can be retrieved and analyzed in groups using Jupyter notebooks, generating results which are ready for publications. This data management server allows to produce reproducible results ready to be shared with the community, thereby promoting open science.