Image analysis platform

With the population imaging platform, users can easily share and run image-analysis pipelines on medical images. The medical images can be stored in a central XNAT database. Then, users can query data and available pipelines, select pipelines for processing and inspect intermediate and final results.

What is BBMRI-NL’s Image analysis platform? 

The Image Analysis Platform is a service that facilitates running medical image-analysis pipelines on medical images. It provides the user with an XNAT database for storing images, an easy-to-use platform to run pipelines on these images to calculate imaging biomarkers, and a viewer to inspect intermediate and final results. Image analysis pipelines can be shared for running at your own institute or on national supercomputing facilities. The tools used for processing are provided under an open-source license (Apache 2.0).

The following tools are provided:

  1. Fastr (http://fastr.readthedocs.io): Pipeline engine and processing manager. Fastr supports visualization of the pipelines and their progress in PIM. This tool is developed by the Erasmus MC in BBMRI WP3 and WP5.
  2. PIM (https://gitlab.com/bbmri/pim_v2): Web framework for online Pipeline Inspection and Monitoring. Exposes a REST API for being interoperable with pipeline engines. This tool is developed by LUMC in BBMRI WP3.
  3. StrongR (https://gitlab.com/bbmri/infra/strongr): StrongR enables truly elastic scaling of batch processes in the cloud, so pipeline frameworks can use all advantages of cloud computing. It will create worker nodes when processing power is required and shut them down if not. StrongR can interface with multiple cloud providers, allowing users to do processing in the cloud of their choice. StrongR is developed by Erasmus MC in BBMRI WP3.

Why the Image Analysis platform?

Biomedical imaging plays a key role in health research, [in establishing diagnostic biomarkers, determine the molecular processes involved, understanding causal pathways, and identify high risk groups.] The dedicated tools and facilities of BBMRI’s imaging and image processing platform make data collection and analysis more efficient. Researchers can share medical imaging data and image-analysis pipelines (using AI and Deep Learning). This is important for better health care, personalized medicine and prevention.

Intended users
The platform is designed to enable researchers to extract biomarkers from imaging data.

Current Status of BBMRI-Imaging
The Image Analysis platform is constantly being developed

Images can be stored in the TraIT XNAT.

Eleven pipelines have been implemented:

  1. Population based statistical analysis of brain structures (LUMC)
  2. Whole brain segmentation (EMC)
  3. Tissue segmentation (EMC)
  4. White-matter-lesion segmentation (EMC)
  5. Brain-structure segmentation (EMC)
  6. Epi-cardial-fat segmentation (EMC)
  7. Intracranial Volume (UMCU)
  8. DTI measures (UMCU)
  9. Subcortical segmentation (Donders Institute)
  10. Independent Connectivity Parcellation (Donders Institute)
  11. Spatial Trend Surface Models (Donders Institute)

How to start using BBMRI-Imaging?

Contact the platform leader or one of the WP leaders.

References and links

www.populationimaging.eu

https://trait.health-ri.nl/trait-tools/xnat

https://trait.topdesk.net/

https://gitlab.com/bbmri

For more information on BBMRI-Imaging contact:

BBMRI-NL’s Population Imaging Platform leader: Marcel Koek: m.koek@erasmusmc.nl or the TraIT Service Desk

BBMRI-NL WP leaders:

Aad van der Lugt: a.vanderlugt@erasmusmc.nl

Wiro Niessen: w.niessen@erasmusmc.nl

Hilleke Hulshoff Pol: h.e.hulshoff@umcutrecht.nl