Tools and guardrails for sharing AI resources at scale
“ATLAS makes it easy for radiologists, researchers, and developers to find AI models and datasets that meet their needs,” said Charles E. Kahn, Jr., MD, a member of the RSNA AI Committee and the Radiology AI Data Standards Subcommittee and editor of ATLAS. Radiology: Artificial Intelligence. “This site helps make AI resources more discoverable and interoperable.”
The platform also provides tools to help create AI index cards, making it easy to share information. ATLAS Card Creator provides templates for creating ATLAS cards. It also features an AI extraction tool that can extract information from existing documents and pre-populate cards for submission.
ATLAS websites and repositories will verify your submissions before publishing them.
- Model cards are checked against JSON schema and live URLs are validated.
- Searchable interface and API: Users can browse and retrieve model cards through an easy-to-use web interface or integrate with the system using an API.
- Ontology-driven indexing: Cards are tagged with RadLex and RSNA content codes for accurate classification.
- Digital Object Identifier (DOI) assignment: Each issued card is given a DOI to ensure traceability and citation.
The Radiology Ontology for AI Datasets, Models, and Projects (ROADMAP) provides a managed terminology of metadata that describes AI models and datasets. The ATLAS data schema and ROADMAP ontology are maintained by a panel of imaging AI experts.
“ATLAS uses widely used vocabularies such as SNOMED and RadLex to index content,” said Dr. Khan. “This site incorporates language from ROADMAP, a newly released set of descriptors from RSNA’s Radiology AI Data Standards Committee to help standardize information about AI resources.”
The imaging AI community is encouraged to publish ATLAS cards for AI models and datasets that they would like to share. Submitting a card ensures that your research is discoverable by the global medical imaging and radiology community, easier to understand and evaluate, and well-positioned for broader collaboration and action.‑Use of the world.
For more information
Learn more about ATLAS today.
Explore additional RSNA AI resources, including datasets and tools for model development.
Explore peer-reviewed research on imaging AI. Radiology: Artificial Intelligence.
See how RSNA is driving innovation and benchmarking through the AI Challenge.
read ahead RSNA News Stories about data standards and interoperability:
