Anonymize DICOM data

This example is a starting point to anonymize DICOM data.

It shows how to read data and replace tags: person names, patient ID, optionally remove curves and private tags, and write the results in a new file.

Anonymize a single file

# authors : Darcy Mason
#           Guillaume Lemaitre <g.lemaitre58@gmail.com>
# license : MIT

import tempfile

from pydicom import examples

ds = examples.mr

for keyword in ["PatientID", "PatientBirthDate"]:
    print(ds.data_element(keyword))

(0010,0020) Patient ID                          LO: '4MR1'
(0010,0030) Patient's Birth Date                DA: ''

We can define a callback function to find all tags corresponding to a person names inside the dataset. We can also define a callback function to remove curves tags.

def person_names_callback(ds, elem):
    if elem.VR == "PN":
        elem.value = "anonymous"


def curves_callback(ds, elem):
    if elem.tag.group & 0xFF00 == 0x5000:
        del ds[elem.tag]

We can use the different callback function to iterate through the dataset but also some other tags such that patient ID, etc.

ds.PatientID = "id"
ds.walk(person_names_callback)
ds.walk(curves_callback)

pydicom allows to remove private tags using remove_private_tags method

ds.remove_private_tags()

Data elements of type 3 (optional) can be easily deleted using del or delattr.

if "OtherPatientIDs" in ds:
    delattr(ds, "OtherPatientIDs")

if "OtherPatientIDsSequence" in ds:
    del ds.OtherPatientIDsSequence

For data elements of type 2, this is possible to blank it by assigning a blank string.

tag = "PatientBirthDate"
if tag in ds:
    ds.data_element(tag).value = "19000101"

Finally, this is possible to store the image

for keyword in ["PatientID", "PatientBirthDate"]:
    print(ds.data_element(keyword))

path = tempfile.NamedTemporaryFile().name
ds.save_as(path)

(0010,0020) Patient ID                          LO: 'id'
(0010,0030) Patient's Birth Date                DA: '19000101'