We’re delighted to announce the launch of the SHIELD-CVD project, an exciting collaboration between the BHF Data Science Centre and Public Health Scotland.
At the heart of SHIELD-CVD (ScottisH Medical Imaging Dataset with Evaluation of Linked Data for CardioVascular Disease) is the Scottish Medical Imaging dataset (SMI) – one of the largest and most comprehensive medical imaging resources in the world, containing over 57 million imaging exams. While this dataset represents an unparalleled opportunity for advancing cardiovascular imaging research, its complexity and scale can make it challenging to use.
The SHIELD-CVD project aims to change that. By providing clear details about the dataset, streamlining access and use, and building reusable data science tools, we will make the SMI more accessible, discoverable, and user-friendly for researchers. Our goal is to help more researchers explore imaging data and carry out studies to improve cardiovascular health for patients.
The project is led by Professor Michelle Williams, Associate Director of the BHF Data Science Centre, working closely with Public Health Scotland. Together, the team is committed to building not only a more usable dataset, but also a sustainable framework for future cardiovascular research.
Now access to the data has been approved, we’re ready for the crucial next stage: data curation and the creation of reusable tools and analysis code to benefit the entire research community.
This work has the potential to speed up progress across cardiovascular imaging research, from understanding how diseases develop to improving population health through early detection and prevention strategies. Importantly the code and insights we develop as part of this project will be transferable to other imaging collections, in the United Kingdom and around the world.
We’re also building a SHIELD-CVD community for people who are using or are interested in using this dataset. Get in touch and let us know what support would help you kickstart your research and please join us for our next webinar on 13 November 2025.
