Project Leads:

Prof. Dr. Marc Dewey (Charité), Prof. Dr. Felix Balzer (Charité), Prof. Dr. Dagmar Krefting (University Medical Center Göttingen), Prof. Dr. Martin Gersch (FUB)

Project Team Members:

Prof. Dr. Daniel Fürstenau (FUB & ECDF), Associate Professor Dr. Stefan Klein (Erasmus MC Rotterdam), Maria Bosserdt, PhD, Melanie Estrella, PhD, Kerstin Rubarth (Charité), James Philip Bowden (UMG), Tim Kilgus (FUB)

Project Start:

05/2023 – 04/2025

Project Description:

There are five major challenges associated with sharing data from clinical imaging studies, which are closely linked to the FAIR Guiding Principles for scientific data management (Findability, Accessibility, Interoperability, Reusability). Here, however, they are grouped into five work packages (WP 1–5) in order to highlight the aspects most relevant to the project.

WP1 – Stakeholder Acceptance
Sharing medical images across different institutions has great potential to advance medical care. However, stakeholders such as patients, participating physicians and institutions, study sponsors, and policymakers may have relevant concerns that need to be addressed in order to gain their acceptance of image sharing.

WP2 – Use and Access
Within the medical imaging community, there is a remarkable cultural shift towards a greater willingness to share data. One important remaining challenge, however, is the implicit potential for re-identification in medical images. In addition, intellectual property issues further complicate the situation.

WP3 – Data Quality
Data quality is crucial to making data suitable and useful for sharing. Meaningful reuse of shared data requires the explicit disclosure of all information relevant to their interpretation, particularly image acquisition parameters, in a generally understandable format.

WP4 – IT Resources
Medical imaging typically generates very large datasets that are highly heterogeneous compared with more commonly shared genomic data. In addition, the nature of imaging data requires substantial storage and computing capacities that must be flexible and scalable in order to integrate the necessary analytical tools.

WP5 – Sustainability
Due to the large and heterogeneous volumes of data, providing an infrastructure for sharing scientific imaging data initially requires considerable investment. Our intention is to establish such an infrastructure for academic use, which requires a sustainability model. It is therefore important to consider costs and benefits beyond the duration of the project.

Our Aim
Our intention is therefore to create such an infrastructure, supported by a sustainability model that takes costs and benefits beyond the project period into account. In addition, gender aspects as well as societal and ethical implications will be identified. All concepts will be aligned with current national and international developments in interoperable data exchange and with the Medical Informatics Initiative in order to enable future integration with existing structures.

Expected Knowledge Gain:
Upon completion of the preliminary project, we will be able to improve the long-term preservation of data from imaging studies by establishing an infrastructure project that meets the requirements of the national and international scientific community. A sustainability plan for this infrastructure will be implemented at the institution's site to ensure efficient, accessible, and long-term use of research data from randomized imaging studies. This will contribute to scientific discoveries, digital medicine, and improved patient outcomes.

GUIDE-IT Pilot
We are currently working on a GUIDE-IT pilot project to test data sharing using data from the DISCHARGE study (coordinated by Marc Dewey's team at Charité) and the SCOT-HEART study (coordinated by Prof. Dr. Michelle Williams' team at the University of Edinburgh). Prof. Williams is also Associate Director of the British Heart Foundation Data Science Centre. Both studies investigate the role of coronary CT angiography in the diagnosis of coronary artery disease and were published in the New England Journal of Medicine (DISCHARGE study in NEJM, SCOT-HEART study in NEJM).

For more information, please visit the GUIDE-IT website.