orthanc-tools 0.17.17

Python Orthanc Tools

python-orthanc-tools

A set of python tools to ease Orthanc scripting.

Functionalities are very limited now ! Backward compat will break a lot in the near future !

Installation

pip3 install orthanc-tools

cloning an Orthanc to another

The cloners copies everything that is currently in the source Orthanc into the destination Orthanc and, once this is done continues the cloning process for every DICOM instance that is received by the source.

from a python script:

from orthanc_tools import OrthancCloner, ClonerMode
from orthanc_api_client import OrthancApiClient

orthanc_a = OrthancApiClient('http://localhost:8042', user='orthanc', pwd='orthanc')
orthanc_b = OrthancApiClient('http://localhost:8043', user='orthanc', pwd='orthanc')

cloner = OrthancCloner(source=orthanc_a, destination=orthanc_b)
cloner.execute(existing_changes_only=False)

# if the destination is declared as a peer:
cloner = OrthancCloner(source=orthanc_a, destination_peer='orthanc-b', mode=ClonerMode.TRANSFER)
cloner.execute(existing_changes_only=False)

from a shell:

python3 -m orthanc_tools.orthanc_cloner --source_url=http://192.168.0.10:8042 --source_user=user --source_pwd=pwd --dest_url=http://192.168.0.10:8042 --dest_user=user --dest_pwd=pwd --run_only_at_night_and_weekend=true --night_start_hour=19 --night_end_hour=6

or, inside a docker-compose file:

services:
    orthanc-cloner:
        image: orthancteam/python-orthanc-tools:0.6.0
        volumes: ["orthanc-cloner:/status"]
        environment:
            TZ: "Etc/UTC"
            RUN_ONLY_AT_NIGHT_AND_WEEKEND: "true"
            NIGHT_START_HOUR: "15"
            NIGHT_END_HOUR: "6"
            SOURCE_URL: "http://orthanc-a:8042"
#            SOURCE_USER: "user"
#            SOURCE_PWD: "pwd"
            DEST_URL: "http://orthanc-b:8042"
#            DEST_USER: "user"
#            DEST_PWD: "pwd"
            MODE: "Default"
            PERSIST_STATE_PATH: "/status/status.txt"
            WORKER_THREADS_COUNT: "6"
#            VERBOSE_ENABLED: "true"
            ERROR_FOLDER_PATH: "/status"
            MAX_RETRIES: "3"
        entrypoint: python -m orthanc_tools.orthanc_cloner
volumes:
    orthanc-cloner:  

OrthancCloner performance

Here are a set of measures performed during a long transfer between 2 VMs running on Azure using OrthancCloner v 0.6.3.

The source Orthanc (v1.9.0) is running on a 4 vCPU VM with 16GB RAM. Postgresql is running on the same VM and DICOM files are stored on data disks are attached to the VM.

The destination Orthanc (v1.11.2) is running on a 4 vCPU VM with 16GB RAM. It is using a flexible managed Postgresql server and an object storage to store DICOM files.

Cloner Mode	WorkersThreadCount	throughput [GB/h]	throughput [instances/h]
Default	12	38	?
Default	18	67	142.000
Default	24	66	160.000
Transfer, Transfers.Threads=6, ConcurrentJobs=2	3	20	?
Transfer, Transfers.Threads=6, ConcurrentJobs=12	6	15	?

import files from a folder from a Docker container

$ docker exec -it xxxx bash

/# pip3 install orthanc-tools

/# python3 -m orthanc_tools.orthanc_folder_importer --folder_path=/import --url=http://localhost:8042 --user=test --password=test --skip_extensions=.cne,.bmp,.ini --worker_threads_count=5

Implement a simple forwarder

The forwarder simply forwards the content of an Orthanc to another DICOM destination and then, deletes the instances. This is usefull for, e.g. implementing an Inbox in front of a PACS that does some IngestTranscoding and/or applies sanitization in a lua script or a python plugin.

from a shell:

python3 -m orthanc_tools.orthanc_forwarder --source_url=http://192.168.0.10:8042 --source_user=user --source_pwd=pwd --destination=target_modality_alias --trigger=StableStudy

migrate DICOM Data from a modality to another

More info in the PacsMigrator class

$ docker exec -it xxxx bash

/# pip3 install orthanc-tools

/# python3 -m orthanc_tools.pacs_migrator --url=http://localhost:8042 --user=user --password=pwd --destination_modality=orthanc-debug --from_study_date=20000101 --to_study_date=20191231 --run_only_at_night_and_weekend --night_start_hour=18 --night_end_hour=6

compare DICOM Data found in Orthanc and in a remote modality

Running in a Docker environment:

$ docker run -d --name comparator --network=mysetup_default python:3.9 bash -c "pip3 install orthanc-tools && python3 -u -m orthanc_tools.orthanc_comparator --level=Instance --url=http://pacs-2022:8042 --modality=pacs-2017 --from_study_date=20220201 --to_study_date=20220302 --transfer_missing_to_modality --ignore_missing_from_orthanc --run_only_at_night_and_weekend --night_start_hour=19 --night_end_hour=6"

uploading a Test DB in Orthanc

The OrthancTestDbPopulator generates test images and uploads them in Orthanc. All images have only 4 pixels and take a minimum amount of space on disk. By default, the generator always generates the same data, use a different seed if you need variation.

From a shell:

python3 -m orthanc_tools.orthanc_test_db_populator --url=http://192.168.0.10:8042 --user=user --password=pwd --studies=5000 --seed=42

purge old studies from an Orthanc

Allows to clean the Orthanc by deleting the oldest studies according to the labels applied on them.

With that sample, all studies with the LABEL1 and older than 6 weeks will be deleted all studies with the LABEL2 and older than 12 weeks will be deleted.

LABEL1,6
LABEL2,12

The script will be executed every day at 2:30 (24 format!)

python3 -m orthanc_tools.orthanc_cleaner --url=http://localhost:8042 --user=orthanc --password=orthanc --execution_time=2:30 --labels_file_path=./tests/stimuli/labels.csv

Deploy an HL7 server parsing ORM^O01 messages to create and store worklists files in a folder

   hl7-server:
        image: orthancteam/python-orthanc-tools:0.10.0
        ports: ["2575:2575"]
        volumes: ["/worklists:/worklists"]
        restart: unless-stopped
        entrypoint: ["python", "-m", "orthanc_tools.hl7_worklist_server_for_orthanc"]

Then, add this env var to Orthanc:

ORTHANC__WORKLISTS__DATABASE: /var/lib/orthanc/worklists