Building representative and generalizable medical imaging datasets increasingly requires looking beyond the boundaries of any single data repository. Interoperability enables data, tools, and resources to work together across research platforms, creating opportunities to combine complementary information while preserving the value and stewardship of individual data commons. Through adherence to the FAIR principles and the use of interoperable infrastructure, researchers can connect imaging data with clinical, genomic, or outcomes data from the same patients across repositories, or bring together similar data types from multiple sources to create larger and more representative cohorts.

The first section, examines the practical benefits that interoperability can provide throughout the research lifecycle, from cohort identification and study design to model development, validation, and deployment.

The second section, explores how interoperability can support representativeness through complementary and distributed data resources. Using practical examples, it illustrates different ways that data from multiple repositories can be combined to enable new research opportunities and richer analyses. This section includes an interactive Jupyter notebook.