Installing OpenREM

Install OpenREM

pip install openrem

Will need ``sudo`` or equivalent if installing on linux without using a virtualenv

Install pynetdicom (edited version)

Pynetdicom is used for the DICOM Store SCP and Query Retrieve SCU functions. See Direct from modalities for details.

pip install


You must install the pynetdicom package from the link above - the version in pypi or the newer versions in GitHub won’t work with the current version of OpenREM. Future versions of OpenREM will be adapted to work with pynetdicom3 and pydicom>1.0.

If you are using the latest version of pip you will get error messages including the phrase Failed building wheel for pynetdicom - it should be ok to ignore this message as long as you end up with the message Successfully installed pynetdicom-0.8.2b2


Locate install location

  • Ubuntu linux: /usr/local/lib/python2.7/dist-packages/openrem/
  • Other linux: /usr/lib/python2.7/site-packages/openrem/
  • Linux virtualenv: vitualenvfolder/lib/python2.7/site-packages/openrem/
  • Windows: C:\Python27\Lib\site-packages\openrem\
  • Windows virtualenv: virtualenvfolder\Lib\site-packages\openrem\

There are two files that need renaming:

  • openremproject/ to openremproject/
  • openremproject/ to openremproject/



Windows notepad will not recognise the Unix style line endings. Please use an editor such as Notepad++ or Notepad2 if you can, else use WordPad – on the View tab you may wish to set the Word wrap to ‘No wrap’


In, always use forward slashes and not backslashes, even for paths on Windows systems.

The directories in this file must already exist - OpenREM will not create them for you.



SQLite is great for getting things running quickly and testing if the setup works, but is not recommended for production use.

We recommend using PostgreSQL as it is the best supported database for Django, and the only one for which the median value will be calculated and displayed in OpenREM charts. Alternatively, other databases such as MySQL/MariaDB, Oracle, and some others with lower levels of support can be used.

There are some further guides to setting up PostgreSQL – see Databases

If you are using SQLite:

'ENGINE': 'django.db.backends.sqlite3',
  • Linux example: 'NAME': '/home/user/openrem/openrem.db',
  • Windows example: 'NAME': 'C:/Users/myusername/Documents/OpenREM/openrem.db',

If you are using PostgreSQL:

'ENGINE': 'django.db.backends.postgresql_psycopg2',
'NAME': 'openremdb',
'USER': 'openremuser',
'PASSWORD': 'openrem_pw',

Location for imports and exports

Csv and xlsx study information exports and patient size csv imports are written to disk at a location defined by MEDIA_ROOT.

The path set for MEDIA_ROOT is up to you, but the user that runs the webserver must have read/write access to the location specified because it is the webserver than reads and writes the files. In a debian linux, this is likely to be www-data for a production install. Remember to use forward slashes for the config file, even for Windows.

Linux example:

MEDIA_ROOT = "/var/openrem/media/"

Windows example:

MEDIA_ROOT = "C:/Users/myusername/Documents/OpenREM/media/"

Secret key

Generate a new secret key and replace the one in the file. You can use for this.

Allowed hosts

The ALLOWED_HOSTS needs to be defined, as the DEBUG mode is now set to False. This needs to contain the OpenREM server hostname or IP address that will be used in the URL in the web browser. For example:


A dot before a hostname allows for subdomains (eg www.doseserver), a dot after a hostname allows for FQDNs (eg Alternatively, a single '*' allows any host, but removes the security the feature gives you.

Customised date format in xlsx exports

# TODO: check csv situation The default date form at for OpenREM xlsx exports is dd/mm/yyyy. If you wish to customise this, uncomment the XLSX_DATE line, for example the standard US date format would be:

XLSX_DATE = 'mm/dd/yyyy'

Log file

There are two places logfiles need to be configured - here and when starting Celery. The logs defined here capture most of the information; the Celery logs just capture workers starting and tasks starting and finishing.

Configure the filename to determine where the logs are written. In linux, you might want to send them to a sub-folder of /var/log/. In this example, they are written to the MEDIA_ROOT; change as appropriate:

import os
logfilename = os.path.join(MEDIA_ROOT, "openrem.log")
qrfilename = os.path.join(MEDIA_ROOT, "openrem_qr.log")
storefilename = os.path.join(MEDIA_ROOT, "openrem_store.log")
extractorfilename = os.path.join(MEDIA_ROOT, "openrem_extractor.log")

LOGGING['handlers']['file']['filename'] = logfilename          # General logs
LOGGING['handlers']['qr_file']['filename'] = qrfilename        # Query Retrieve SCU logs
LOGGING['handlers']['store_file']['filename'] = storefilename  # Store SCP logs
LOGGING['handlers']['extractor_file']['filename'] = extractorfilename  # Extractor logs

If you want all the logs in one file, simply set them all to the same filename.

In the settings file, there are simple and verbose log message styles. We recommend you leave these as verbose:

LOGGING['handlers']['file']['formatter'] = 'verbose'        # General logs
LOGGING['handlers']['qr_file']['formatter'] = 'verbose'     # Query Retrieve SCU logs
LOGGING['handlers']['store_file']['formatter'] = 'verbose'  # Store SCP logs
LOGGING['handlers']['extractor_file']['formatter'] = 'verbose'  # Extractor logs

Next, you can set the logging level. Options are DEBUG, INFO, WARNING, ERROR, and CRITICAL, with progressively less logging. INFO is probably a good choice for most circumstances. DEBUG is useful if something is going wrong, but it is quite chatty for routine use!

LOGGING['loggers']['remapp']['level'] = 'INFO'                    # General logs
LOGGING['loggers']['remapp.netdicom.qrscu']['level'] = 'INFO'     # Query Retrieve SCU logs
LOGGING['loggers']['remapp.netdicom.storescp']['level'] = 'INFO'  # Store SCP logs
LOGGING['loggers']['remapp.extractors.ct_toshiba']['level'] = 'INFO'  # Toshiba RDSR creation extractor logs

Finally, if you are using Linux you can set the system to start a new log file automatically when the current one gets to a certain size. The settings described below don’t work with Windows - we’ll try to include Windows settings in the next release. See issue 483 to find out the progress on this!

To activate the ‘rotating’ log function, uncomment the remaining lines by removing the # from the beginning of the lines. For example for the query retrieve logs:

LOGGING['handlers']['qr_file']['class'] = 'logging.handlers.RotatingFileHandler'
LOGGING['handlers']['qr_file']['maxBytes'] = 10 * 1024 * 1024  # 10*1024*1024 = 10 MB
LOGGING['handlers']['qr_file']['backupCount'] = 5  # number of log files to keep before deleting the oldest one

Time zone

Configure the local timezone in order to get correct times in the logfiles. The default timezone is set at ‘Europe/London’. Valid options can be found here:

TIME_ZONE = 'Europe/London'


Configure the local language. Default language is set at us-english. Valid options can be found here:


Toshiba CT RDSR creation

If you need to import data from older Toshiba CT scanners into OpenREM then the following tools need to be available on the same server as OpenREM:

The paths to these must be set in for your system:

# Locations of various tools for DICOM RDSR creation from CT images
DCMTK_PATH = 'C:/Apps/dcmtk-3.6.0-win32-i386/bin'
DCMCONV = os.path.join(DCMTK_PATH, 'dcmconv.exe')
DCMMKDIR = os.path.join(DCMTK_PATH, 'dcmmkdir.exe')
JAVA_EXE = 'C:/Apps/doseUtility/windows/jre/bin/java.exe'
JAVA_OPTIONS = '-Xms256m -Xmx512m -Xss1m -cp'
PIXELMED_JAR = 'C:/Apps/doseUtility/pixelmed.jar'
PIXELMED_JAR_OPTIONS = '-Djava.awt.headless=true com.pixelmed.doseocr.OCR -'

The example above is for Windows. On linux, if you have installed the Offis DICOM toolkit with sudo apt install dcmtk or similar, you can find the path for the configuration above using the command which dcmconv. This will be something like /usr/bin/dcmconv, so the DCMTK_PATH would be '/usr/bin and the DCMCONV would be os.path.join(DCMTK_PATH, 'dcmconv'). Similarly for DCMMKDIR and JAVA_EXE, which might be /usr/bin/java. The pixelmed.jar file should be downloaded from the link above, and you will need to provide the path to where you have saved it.

Create the database

In a shell/command window, move into the openrem folder:

  • Ubuntu linux: cd /usr/local/lib/python2.7/dist-packages/openrem/
  • Other linux: cd /usr/lib/python2.7/site-packages/openrem/
  • Linux virtualenv: cd virtualenvfolder/lib/python2.7/site-packages/openrem/
  • Windows: cd C:\Python27\Lib\site-packages\openrem\
  • Windows virtualenv: cd virtualenvfolder\Lib\site-packages\openrem\

Create the database:

python makemigrations remapp
python migrate
python showmigrations

The last command will list each Django app migrations. Each should have a cross inside a pair of square brackets something like below:

 [X] 0001_initial
 [X] 0001_initial
 [X] 0002_alter_permission_name_max_length
 [X] 0003_alter_user_email_max_length
 [X] 0004_alter_user_username_opts
 [X] 0005_alter_user_last_login_null
 [X] 0006_require_contenttypes_0002
 [X] 0001_initial
 [X] 0002_remove_content_type_name
 [X] 0001_initial
 [X] 0001_initial
 [X] 0001_initial

Finally, create a Django super user:

python createsuperuser

Answer each question as it is asked – this user is needed to set up the other users and the permissions.

Add the median database function: PostgreSQL databases only

Rename the file




and then run

python migrate

This command runs the migration file, and will display the text Applying remapp.0002_0_7_fresh_install_add_median... OK, indicating that the median function has been added.

Start all the services!

You are now ready to start the services to allow you to use OpenREM - go to Start all the services to see how!