Azure Health Data Services De-identification Service API
A stateless REST API service that uses machine learning to automatically identify and process 27 PHI entity types (including HIPAA's 18 identifiers) in unstructured clinical text. Supports three operations: TAG (identify and label PHI), REDACT (replace PHI with entity type placeholder), and SURROGATE (replace PHI with realistic pseudonyms for consistent surrogation within a batch). Offers synchronous single-document and asynchronous batch endpoints for Blob Storage processing. Service limits: 50 KB per request, 10,000 documents per job, 2 MB per document, throttled at 1 MB per 5 seconds or 100 requests per 5 seconds. Includes 50 MB free monthly allowance.
Documentation
Documentation
https://learn.microsoft.com/en-us/azure/healthcare-apis/deidentification/overview
Authentication
https://learn.microsoft.com/en-us/azure/healthcare-apis/authentication-authorization
Specifications
Other Resources
REST API Reference
https://learn.microsoft.com/en-us/rest/api/health-dataplane/deidentify-text/deidentify-text
SDK Overview
https://learn.microsoft.com/dotnet/api/overview/azure/Health.Deidentification-readme
Service Limits
https://learn.microsoft.com/en-us/azure/healthcare-apis/deidentification/overview#input-requirements-and-service-limits
Pricing
https://azure.microsoft.com/en-us/pricing/details/health-data-services/