Federated European infrastructure for cancer images data

Outcomes and deliverables

The following elements shall be delivered by the selected project:

• Deployment of interoperable, FAIR compliant and secure infrastructure and data governance enabling a sustainable cross-border connection of cancer image data sources. The infrastructure will serve clinicians, researchers and innovators in combination with technologies and tools necessary for data analysis.
• Support for actions aiming to create or extend cancer image data sources (including image annotation) and/or to adapt existing (legacy) data data.
• Clinical validation and use of high-end computing and simulation resources for data analysis, and a federated AI learning system, as well as deployment of a secure authentication system.
• Business model including uptake strategy explaining the motivation and incentives for all stakeholders at the different levels (national, European, global) to support the data infrastructure towards its sustainability, including data controllers (hospitals/municipalities, research institutes, patients), data users (clinicians, researchers, policy makers, companies), service providers (e.g. IT industry, medical imaging companies), healthcare systems and public authorities at large.

The aim is to establish and deploy a pan-European digital infrastructure facilitating access to cancer images and related patient data in full compliance with the applicable data protection requirements. It will be used by clinicians, researchers and innovators with the ultimate aim of more precise and faster clinical decision-making, diagnostics, treatments and predictive medicine that will benefit citizens, patients, healthcare systems and the overall economy. The resulting data infrastructure will be inter-operable with the other components of the European health data space, in particular the genomic data infrastructure (see topic 2.2.1.7.1). It contributes to the Europe's Beating Cancer Plan and to the cancer mission under Horizon Europe, while building on relevant research projects under Horizon 2020. In order to maximise the societal benefits of health data use, the data space should be supported by advanced IT tools and capacities, e.g. AI and HPC. Moreover, the measure will support creation of new cancer image datasets and further extension of the existing ones.

This action will support the deployment of the infrastructure needed to link and explore fragmented European databases of medical images of different types of cancer, with a solid governance and clear business model for ingestion of data and its exploitation by public and private organisations in Member States and associated countries (controllers of data), industry and innovators. The infrastructure should consist of a central coordination entity, a federated network of FAIR-conform connected data sources, a platform enabling data discovery, querying and capability to access appropriate computing capacities for distributed data analysis. The action may combine existing data sources and support creation of new data sets. To ensure maximum data protection, identifiable data will in principle be analysed using distributed data analysis and AI learning techniques, while fully taking into account the applicable data protection requirements and the EU’s international obligations. Central storage may take place if the data are anonymised and in accordance with applicable legal requirements. The project will be in full compliance with the relevant legal, ethical, quality and interoperability requirements and standards.

The infrastructure comprising different data sources must be based on common interoperability mechanisms. Therefore, the implementation of the data infrastructure should build on the progress achieved and broad agreements made in the relevant research projects (under Horizon 2020) regarding the relevant interoperability mechanisms such as system architecture, specifications on reference APIs, meta-data, data structure and quality, legal requirements etc. The project should also enable linking and analysing cancer image data together with other types of health data (in particular electronic health records, repositories of molecular and clinical information and genomic-phenotypic data).

The project is expected to engage with patients, citizens, health professionals and other stakeholders to explain that data is used transparently and responsibly, and raise awareness of the expected benefits for European patients and citizens.

The project selected for the deployment of this data infrastructure will have to make provisions for gradually becoming fully compliant with the European Data Spaces Technical Framework. It will also have to coordinate and collaborate with other projects participating in the deployment of the data space and the Data Spaces Support Centre in order to allow integration of existing standards and to ensure interoperability and portability across infrastructure, applications and data.

Furthermore, the project implementing the data infrastructure for cancer images is encouraged to cooperate with Testing and Experimentation Facilities for Health (see topic 2.3.2.2), to define European test and training data sets and to provide support in their establishment.