Tutorial: Use a custom redaction format with the de-identification service

In this tutorial, you:

Usage

Within the de-identification service, we support an operation called Redact that allows you to redact PHI from a text document. You can specify what that redaction looks like using a custom format.

For example:

Specify a custom redaction format

1. Must select Redact as Operation. RedactionFormat is only supported for Redact operation.
2. Pass RedactionFormat parameter within the CustomizationOptions model to the API or Job parameters.

Variables

Redaction format variables refer to special placeholders that can be used to create a custom redaction format.

The following variables are supported:

Type

{type} => patient

    Text = "Hi my name is John Smith"
    RedactionFormat = "<{type}>"

    # Output:
    Hi my name is <patient>

Also supports Upper and Title cases.

    {type} => patient
    {Type} => Patient
    {TYPE} => PATIENT

Length

This variable allows you to create a string matching the length of the PHI.

It duplicates the previous character to match the length of the tagged entity.

    Text = "Hi my name is John Smith"
    RedactionFormat = "*{len}"

    # Output:
    Hi my name is **********

Limits

• The redaction format can be up to 16 characters long.
• Each variable type can only be used once in the format.