SantaClean
A local de-identification tool built to help prepare healthcare data for software development.
The barrier it addresses
A home-care agency needed software to coordinate emergency caregiver shifts. Developing that solution required useful test data, but the agency’s source records contained protected client information.
SantaClean was built to prepare a de-identified copy on the owner’s computer. This gives the project a way to work toward usable development data while keeping the original records and the local identity mapping under the owner’s control.
What the tool does
- Processes CSV, text, and Word documents locally.
- Uses pattern matching, document context, and local models to detect identifiers.
- Replaces or removes identifiers and presents uncertain classifications for human review.
- Runs a separate scan of the cleaned output and produces a local audit record.
The intended handoff is the reviewed, cleaned copy. Original records and the information used to map replacement codes back to identities stay on the owner’s computer.
Current project status
SantaClean is a working release candidate awaiting the owner’s evaluation. It is the current tool; the earlier agency-specific de-identifier was its first iteration.
The scheduling project has a problem definition, solution proposal, staged build plan, and tested deterministic matching prototype. The complete scheduler has not been built. Further integration awaits the owner’s export and final input.
SantaClean supports data preparation and review. Its output still needs to be assessed for the intended use; running the tool alone does not certify that a dataset is suitable to share.
Examine the work
A project walkthrough can cover the original development barrier, the tool’s workflow, and how it fits into the scheduling plan.
Ask about SantaClean →