| ID | product.analytics.deid_pipeline |
|---|---|
| Description | Strips identifiers per Safe Harbor rules. |
| Key | deid_pipeline |
| Type | microservice:service |
| Team | Analytics & Research |
| Owner | analytics@healthcare.example |
| Status | active |
| Technologies | Spark, PostgreSQL, OMOP |
| Attributes | schedule: nightly batch, standard: HIPAA Safe Harbor |
| Tags |
The De-identification Pipeline applies HIPAA Safe Harbor transforms before any row reaches the analytics warehouse.
Block risky publishes
If re-identification risk exceeds policy, the batch fails until compliance approves or transforms adjust.
Could not build diagram for model "example-healthcare".
The Identifier Scrubber handles the 18 HIPAA identifier classes.
The Date Shifter applies per-patient salt so timelines stay internally consistent.
The Re-identification Risk Scorer estimates residual re-id risk before publish.
2 workflows
| Workflow | Model | Participant id | Participant label |
|---|---|---|---|
| Consent update governance.consent-update | example-healthcare | product.analytics.deid_pipeline | De-id pipeline |
| Researcher query governance.researcher-query | example-healthcare | product.analytics.deid_pipeline | De-id pipeline |