Secondary use of radiological imaging data: Vanderbilt's ImageVU approach

Area of Science:

• Medical Informatics
• Radiology Research
• Data Infrastructure

Background:

• Clinical imaging data is crucial for research but often difficult to access for secondary use.
• Existing systems may lack scalability, efficiency, and regulatory compliance for research data sharing.
• There is a need for integrated infrastructure to bridge clinical imaging and research workflows.

Purpose of the Study:

• To develop ImageVU, a scalable research imaging infrastructure.
• To enable secure, efficient, and regulatory-compliant access to clinical imaging data for secondary research.
• To provide a detailed description of ImageVU's components and lessons learned for other institutions.

Main Methods:

• ImageVU integrates clinical imaging data with metadata-driven cohort discovery.
• Key components include a Research PACS, Ad Hoc Backfill Host, Cloud Storage, and De-Identification System.
• Metadata is stored in identified (RD) and de-identified (SD) repositories, with access via the RD Discover web portal.

Main Results:

• ImageVU streamlines cohort discovery and improves imaging data accessibility.
• Processed 12.9 million MRI/CT series from 1.36 million studies (453,403 patients) by December 2024.
• Supported 75 projects, delivering 50 TB of data to 55 investigators, resulting in 66 publications.

Conclusions:

• ImageVU offers a scalable, efficient approach for integrating clinical imaging into research.
• The platform combines institutional data infrastructure with cloud storage and metadata for secure, compliant research access.
• ImageVU facilitates translational research by enabling easier access to imaging data.