6 releases

0.7.0 Apr 25, 2024
0.2.2 Nov 21, 2023
0.2.1 Aug 26, 2023
0.2.0 Jul 23, 2023
0.1.0 Oct 23, 2022

#1595 in Command line utilities

Download history 20/week @ 2024-02-21 1/week @ 2024-02-28 1/week @ 2024-03-06 14/week @ 2024-03-13 17/week @ 2024-03-27 18/week @ 2024-04-03 121/week @ 2024-04-24 6/week @ 2024-05-01

359 downloads per month

MIT/Apache

2.5MB
42K SLoC

DICOM-rs findscu

CratesIO Documentation

This is an implementation of the DICOM Find SCU (C-FIND), which can be used to search for study and patient records in a DICOM archive.

This tool is part of the DICOM-rs project.

Usage

Note that this tool is not necessarily a drop-in replacement for findscu tools in other DICOM software toolkits. Run dicom-findscu --help for more details.

Basic usage includes searching for a study or patient by a certain attribute. The following query/retrieve information models are supported at the moment:

  • -S: Study Root Query/Retrieve Information Model – FIND (default)
  • -P: Patient Root Query/Retrieve Information Model - FIND
  • -W: Modality Worklist Information Model – FIND

There are two non-exclusive ways to specify a DICOM query:

Passing a DICOM query object file

You may optionally start by providing a DICOM query object as a file. There are currently no tools in DICOM-rs to assist in the process of creating these objects, but one can convert DCMTK DICOM data dumps into compatible DICOM query objects, or write these tools yourself.

# query is defined in file
dicom-findscu PACS@pacs.example.com:1045 --study query.dcm

Using the multi-value -q option

Each value is a text of the form «field_path»=«field_value», where:

  • field_path is a data element selector path;
  • and field_value is the respective value or pattern to match against the value of the specified DICOM attribute. It can be empty, which in that case the = may also be left out.

If a path to a DICOM query file is passed, these options will extend or override the query object from the file.

Selector syntax

Simple attribute selectors comprise a single data element key, specified by a standard DICOM tag (in one of the forms (gggg,eeee), gggg,eeee, or ggggeeee) or a tag keyword name such as PatientName. To specify a sequence, use multiple of these separated by a dot (e.g. ScheduledProcedureStepSequence.0040,0020). When writing nested attributes, you currently need to declare the sequence in a separate query option, and only then define the attribute inside it, in this order. See the examples below for clarity.

Examples

# query application entity STORAGE for a study with the accession number A123
dicom-findscu STORAGE@pacs.example.com:1045 --study -q AccessionNumber=A123

# query application entity PACS for patients born in 1990-12-25
dicom-findscu PACS@pacs.example.com:1045 --patient -q PatientBirthDate=19901225

# wild-card query: grab a list of all study instance UIDs
dicom-findscu PACS@pacs.example.com:1045 -S -q "StudyInstanceUID=*"

# retrieve the modality worklist information
# for scheduled procedures where the patient has arrived
dicom-findscu INFO@pacs.example.com:1045 --mwl \
    -q ScheduledProcedureStepSequence \
    -q ScheduledProcedureStepSequence.ScheduledProcedureStepStatus=ARRIVED

Dependencies

~8–19MB
~254K SLoC