Work Package 2

Management Team

Main Objective

The objective of WP2 is to deliver the infrastructure hosting the several petabytes of 3 million whole-slide images (WSIs). The infrastructure ensures: the data register and the corresponding files and associated metadata, and makes the data accessible for research. The infrastructure works closely with data brokering and data analysis software in Bigpicture. The infrastructure is being designed towards long-term sustainability, including extensibility and durability to ensure amenability to future expansion in alignment with sustainability plans. The infrastructure design is coordinated with consortium partners to ensure interoperability with existing tools in academia, industry and healthcare.

Main Tasks 

Phase I – Preparations

Task 2.1 – Infrastructure Requirements gathering. Establishment of initial requirements on infrastructure as understood by academic and industrial partners, with a starting point in existing services for genomics (ELIXIR AAI, REMS, Federated EGA [code, documentation], Beacon) and for biobank discoverability (BBMRI Directory). This task is intended to map out what developments are necessary in order for these services to be used or adapted to support digital pathology and relate them to the proposed high-level architecture.

Task 2.2 – Establish development platform. This task will establish a state-of-the-art systems development platform to support collaborative development of digital pathology adaptations to forks of existing services, including version control, issue tracking, continuous integration and deployment.

Phase II – Pilot system development

Task 2.3 – Pilot system development. In this task, we will establish a pilot system based on initial requirements gathered in task 2.1, to allow pilot data taking and inference in task 2.5. This includes software components and necessary hardware for system integration and testing, as well as planning and implementing production practices and documentation.

Task 2.4 – Security review of pilot systems. Within this task we will perform a security review of the pilot systems in collaboration with independent external cyber security expertise to verify readiness for pilot data taking and inference in task 2.5.

Phase III – Refinements

Task 2.5 – Pilot operation. This task will focus on targeted data incorporation in the repository in collaboration with the pilot data contributors in WP3 through user interfaces developed in WP4 based on APIs developed in Task 2.3.

Task 2.6 – Iterative development. This task involves regular requirement dialogue, structured similarly as in Task 2.1. The dialogue will include the Bigpicture consortium and WPLs and nominated stakeholder representatives as necessary to progressively refine requirements on infrastructure.

Task 2.7 – Regular security review. Annual security review of components where changes have been made, and penetration testing of the integrated system in collaboration with independent external cyber security expertise.

Phase IV – Scaleup

Task 2.8 – Large scale archive operations. Broad engagement with wide range of Bigpicture data- and inference model contributors and users through interfaces developed in WP4, expecting to receive data at approximately twice the rate as compared to the pilot phase.

Bigpicture: Meet Maarten Vergauwen

In this video, Maarten Vergauwen, WP2 Lead EFPIA, explains his role in the project.