[ivory-search id="3463" title="Default Search Form"]
Search
Register here
Search
Type below to search:
OnlineDiscovery - AI and Health Badge available

Data standards for health AI: Benchmarking, metadata and federated data discovery

  • Register

    * Register (or log in) to the Neural Network to add this session to your agenda or watch the replay

    • Date
    27 March 2026
    Timeframe
    15:00 - 16:00 CET Geneva
    Duration
    60 minutes
    • Days
      Hours
      Min
      Sec

    This webinar presents the data infrastructure work of the Global Initiative on AI for Health (GI-AI4H), the successor to the ITU/WHO Focus Group on AI for Health (FG-AI4H). Since 2018, the Open Code Initiative (OCI) has developed a benchmarking and assessment platform for health AI solutions, engaging over 40 contributors across five continents. Building on this foundation, the GI-AI4H is now advancing three critical data infrastructure pillars.

    First, the OCI evaluation platform, a key FG-AI4H deliverable, supports end-to-end assessment of health AI algorithms under regulatory guidelines. Second, BioCroissant, a life sciences extension of the MLCommons Croissant metadata standard (v1.1, Feb 2026), enables standardized, machine-readable descriptions of biomedical datasets, linking to domain-specific ontologies for improved discoverability and interoperability. Third, a federated data catalogue and marketplace concept that allows health datasets to remain at their source while being discoverable and accessible for AI development, addressing data sovereignty and privacy challenges in global health.

    This session will equip participants with an understanding of how these interconnected components form a comprehensive data ecosystem for trustworthy health AI.

     

    Session Objectives:

    By the end of this session, participants will be able to:

    • Describe the goals and structure of the ITU-WHO-WIPO Global Initiative on AI for Health.
    • Describe the role of the OCI evaluation platform in standardized health AI benchmarking.
    • Explain how the BioCroissant metadata standard extends Croissant to support biomedical dataset description, discoverability, and interoperability across ML platforms.
    • Differentiate between centralized and federated approaches to health data cataloguing, and articulate the benefits of a federated marketplace for data sovereignty and privacy.
    • Evaluate how integrated data standards (metadata, cataloging, benchmarking) contribute to building trustworthy and equitable AI for health solutions globally.
    • Apply knowledge of these data infrastructure components to identify opportunities for contributions.

    Recommended Mastery Level/Prerequisites

    Beginner to Intermediate. No prior technical expertise required. A general familiarity with AI/ML concepts and interest in health data governance is helpful. The session is suitable for health professionals, data scientists, policymakers, regulators, and researchers interested in AI for health standardization.

    Share this session with your network
    • Session starts in
      Days
      Hours
      Min
      Sec
    Add to Calendar
    Discover more from programme stream
    Discovery - AI and Health
    Data standards for health AI: Benchmarking, metadata and federated data...

    Data standards for health AI: Benchmarking, metadata and federated data...

    27 March 2026 - 15:00 to 16:00

    Related stories

    Are you sure you want to remove this speaker?