Medical imaging data is notoriously difficult to manage. Computed Tomography (CT) scans can generate hundreds of DICOM files, each with its own scanner-specific metadata fields, quirks, and inconsistencies that change dramatically over time and with new equipment. For researchers, that complexity often becomes a bottleneck, making it hard to access exactly the scans they need to learn from the data.
Arizona State University’s Center for Evolution and Medicine faced this challenge with the Tsimane Health and Life History Project, a two-decade effort to understand healthy aging in a non-industrialized Bolivian population that exhibited excellent long-term health. The project’s CT imaging dataset grew into a critical resource for studying cardiovascular and cognitive health in aging populations, but getting the right scans to researchers required weeks of manual filtering and custom curation.
Working in partnership with Starfish Storage, Suhail Ghafoor, Information Technology Manager for the Center, built a metadata-driven catalog that transformed unstructured DICOM archives into a structured, searchable, researcher-friendly data access system. The solution:
- Normalized inconsistent metadata into standard categories like body part, scan quality, and reconstruction type
- Imported that metadata into Starfish, making it fully queryable.
- Exposed results to researchers through virtual directory structures that offered self-service, storage-efficient, and tailored data access to researchers.
The impact has been dramatic. Requests that once took months now take minutes. Researchers can browse and filter data independently, without requiring IT intervention, and the system runs sustainably on ASU-controlled infrastructure, meeting strict governance and privacy standards.
Most importantly, ASU’s solution is one that any institution facing similar challenges with DICOM data can replicate. By combining metadata normalization with Starfish’s flexible cataloging, querying, data management, and scripting capabilities, researchers and clinical IT teams can implement similar solutions.
The complete story, including the technical details of metadata normalization, directory structures, and links to the open-source scripts that power the pipeline, is now available in our new white paper: How ASU’s Center for Evolution and Medicine Used Starfish to Build a Searchable DICOM Catalog for Global Health Research
2026 Data Breakthrough Awards
Starfish Storage Wins “Data Solution of the Year for Research”
Starfish Storage has been named 2026 Data Solution of the Year for Research by the Data Breakthrough Awards, recognizing its work with Arizona State University to deliver governed, self-service access to decades of CT imaging data and help researchers move faster.
