dicom-anonymization

This repository provides a library and binary for anonymizing (or de-identifying) DICOM files.

[!WARNING] This is a work in progress. Some things may still change and not all things may work as expected yet.

The project prioritizes performance, reliability, safety and ease of use.

Building

cargo build --release

Usage

See documentation on docs.rs.

Python

See the Python bindings README.md.

Library

Installation

To add the library to your project, do this:

cargo add dicom-anonymization

Using default configuration

use std::fs::File;
use dicom_anonymization::Anonymizer;

let file = File::open("tests/data/test.dcm")?;

let anonymizer = Anonymizer::default();
let result = anonymizer.anonymize(file)?;

let output_file = File::create("anonymized.dcm")?;
result.write(output_file)?;

Using custom configuration

use std::fs::File;
use dicom_anonymization::tags;
use dicom_anonymization::Anonymizer;
use dicom_anonymization::actions::Action;
use dicom_anonymization::config::builder::ConfigBuilder;
use dicom_anonymization::processor::DefaultProcessor;

// default configuration can be customized/overridden
let config = ConfigBuilder::default()
    .uid_root("1.2.840.123".parse().unwrap())
    .remove_private_tags(true)
    .tag_action(tags::PATIENT_NAME, Action::Replace { value: "John Doe".into() })
    .tag_action(tags::PATIENT_ID, Action::Hash { length: Some(16) })
    .tag_action(tags::ACCESSION_NUMBER, Action::Hash { length: Some(16) })
    .tag_action(tags::STUDY_DATE, Action::HashDate)
    .tag_action(tags::SERIES_DATE, Action::Remove)
    .tag_action(tags::STUDY_INSTANCE_UID, Action::HashUID)
    .tag_action(tags::SERIES_INSTANCE_UID, Action::HashUID)
    .tag_action(tags::SOP_INSTANCE_UID, Action::HashUID)
    .build();

let processor = DefaultProcessor::new(config);
let anonymizer = Anonymizer::new(processor);

let file = File::open("tests/data/test.dcm")?;
let result = anonymizer.anonymize(file)?;

let mut output = Vec::<u8>::new();
result.write(&mut output)?;

Building configuration from scratch

use dicom_anonymization::tags;
use dicom_anonymization::actions::Action;
use dicom_anonymization::config::builder::ConfigBuilder;

let config_from_scratch = ConfigBuilder::new()
    .uid_root("1.2.840.123".parse().unwrap())
    .remove_private_tags(false)
    .tag_action(tags::PATIENT_NAME, Action::Replace { value: "John Doe".into() })
    // ...more config rules...
    .build();

Binary

Installation

To install the dcmanon binary, do this:

cargo install dicom-anonymization

Usage

anonymize

$ dcmanon anonymize --help
Anonymize DICOM files

Usage: dcmanon anonymize [OPTIONS] --input <INPUT_PATH> --output <OUTPUT_PATH>

Options:
  -i, --input <INPUT_PATH>    Input file ('-' for stdin) or directory
  -o, --output <OUTPUT_PATH>  Output file ('-' for stdout) or directory
  -c, --config <CONFIG_FILE>  Path to config JSON file
  -u, --uid-root <UID_ROOT>   UID root (default '9999')
      --exclude <TAGS>        Tags to exclude from anonymization, e.g. '00100020,00080050'
  -r, --recursive             Recursively look for files in input directory
      --continue              Continue when file found is not DICOM
  -v, --verbose               Show more verbose output
  -d, --debug                 Show debug output
  -h, --help                  Print help
  -V, --version               Print version

Example

dcmanon -i tests/data/test.dcm -o anonymized.dcm

config create

$ dcmanon config create --help
Create a configuration file

Usage: dcmanon config create [OPTIONS]

Options:
  -o, --output <CONFIG_FILE>  Path to save the config file  (‘-’ or omitted → stdout) [default: -]
  -v, --verbose               Show more verbose output
  -d, --debug                 Show debug output
  -u, --uid-root <UID_ROOT>   UID root to use
      --exclude <TAGS>        Tags to exclude from anonymization, e.g. '00100020,00080050'
      --diff-only             Only output the dfferences with the default config
  -h, --help                  Print help

Contributing

We welcome contributions from the community. If you are interested in contributing to the project, please read CONTRIBUTING.md.

License

Licensed under either of

• Apache License, Version 2.0 (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
• MIT License (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.