Skip to main content

Expanded, Multimodal De-Identification Tool Suite

As more analytical applications expand opportunities for discovery, the need for large volumes of clinical data grows. However, security and privacy are critical to accessing this data, particularly when collaboration is involved. Flywheel offers a research solution with support for regulatory-compliant, configurable medical imaging de-identification (MIDI) to streamline data access and accelerate discovery.

A Research Data Platform De-identifies on the Edge

Many established clinical systems are not well-suited for research. Researchers need to view cohorts instead of individuals, collaborate with specific groups of contributors, and most importantly, limit the protected health information (PHI) in their data to comply with regulations and IRB oversight.

A research data platform is a separate system from clinical PACS and VNAs that is dedicated to enabling research on subjects with the minimum protected health information necessary. Flywheel de-identifies data as it is captured, outside of Flywheel’s platform, giving researchers rapid access only to the information they need.

Ensure Regulatory Compliance

Flywheel enables compliance with IRB requirements, HIPAA, GDPR, and 21 CFR Part 11. Customers can set a minimum level of de-identification for all research projects at their institution. Additionally, Flywheel authenticates users via the institution’s authentication service and features a role-based permissions model so only the necessary users can access study data.

Flywheel also supports data and algorithm provenance, approval workflows, and audit trails. To enable re-identification per GDPR, Flywheel can generate a pseudonymization log at the time of data ingestion, which customers can use to re-ID data outside of our platform.

Configure De-identification to Unique Project Needs

Regulations require de-identification of protected health information to the minimum level required for the intended use. However, the minimum level of identification may differ by project. Flywheel integrates de-identification in a way that meets the needs of projects with different requirements while maintaining compliance.

Administrators may set custom de-identification rules that remove PHI from imaging data, EMR and more. These rules can perform a number of actions on PHI data, including:

  • Keeplists
  • Removelists
  • Hashing
  • Replacing
  • Date offsetting
  • Optical character recognition
  • De-facing
  • and more…

Flywheel features de-identification through all of its capture tools, including its Edge Connectors for research instruments, Bulk Import via command line interface, as well as through the zero-footprint Web Uploader. 

De-identification rules can be further specified to a group of projects or a specific project. Administrators simply attach de-identification rules to the project in the user interface. 


Flywheel's de-identification options unleash the value of clinical data for your researchers. Please reach out to with any questions about Flywheel’s support for de-identification and regulatory compliance.