Welcome to OpenREM’s documentation!¶

Databases¶

During the installation process, you will need to install a database. For testing only, you can use the built in SQLite3 database, but for production use you will need a production grade database. This is covered in the Pre-installation preparations documentation, but as you will probably want to find the database instructions again, the links are repeated here.

sudo apt-get install postgresql libpq-dev

pip install psycopg2

sudo nano /etc/postgresql/9.4/main/pg_hba.conf

local    all            all                         md5

sudo service postgresql restart

sudo service postgresql stop
mkdir /var/openrem/database
sudo cp -aRv /var/lib/postgresql/9.4/main /var/openrem/database/
sudo nano /etc/postgresql/9.4/main/postgresql.conf

data_directory = '/var/lib/postgresql/9.4/main'

data_directory = '/var/openrem/database/main'

sudo service postgresql start

sudo -u postgres createuser -P openremuser

sudo -u postgres createdb -T template1 -O openremuser -E 'UTF8' openremdb

nano openremproject/local_settings.py

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

sudo -u postgres pg_dump openremdb > /path/to/backup.bak

tar -czf backup.bak.tar.gz backup.bak

#! /bin/bash
rm -rf /path/to/db/backups/*
PGPASSWORD="openrem_pw" /usr/bin/pg_dump -Uopenremuser openremdb > /path/to/db/backups/openrem.bak

sudo -u postgres createdb -T template0 new_openremdb_name
sudo -u postgres psql new_openremdb_name < /path/to/db/backups/openrem.bak

-- Start the PostgreSQL console
sudo -u postgres psql

-- List users
\du

-- List databases
\l

-- Exit the console
\q

pip install psycopg2-2.6.1-cp27-cp27m-win32.whl
# or if you have the 64-bit version
pip install psycopg2-2.6.1-cp27-cp27m-win_amd64.whl
# adjusting the version number appropriately

C:\Program Files\MySQL\MySQL Server 5.6\bin\mysql.exe -u root -p -e "CREATE USER 'backup'@'localhost' IDENTIFIED BY 'backup'";

C:\Program Files\MySQL\MySQL Server 5.6\bin\mysql.exe -u root -p -e "GRANT ALL PRIVILEGES ON openremdatabasemysql .* TO 'backup'@'localhost'";

C:\Program Files\MySQL\MySQL Server 5.6\bin\mysql.exe -u root -p -e "GRANT CREATE ON *.* TO 'backup'@'localhost'";

C:\Program Files\MySQL\MySQL Server 5.6\bin\mysql.exe -u root -p -e "FLUSH PRIVILEGES";

C:\Program Files\MySQL\MySQL Server 5.6\bin\mysqldump.exe -ubackup -pbackup openremdatabasemysql > backup.sql

C:\Program Files\MySQL\MySQL Server 5.6\bin\mysql.exe -ubackup -pbackup -e "CREATE DATABASE openremdatabasemysql";

C:\Program Files\MySQL\MySQL Server 5.6\bin\mysql.exe -ubackup -pbackup openremdatabasemysql < backup.sql

@echo off
For /f "tokens=1-4 delims=/ " %%a in ('date /t') do (set mydate=%%c-%%b-%%a)
For /f "tokens=1-2 delims=/:" %%a in ('time /t') do (set mytime=%%a%%b)

"C:\Program Files\MySQL\MySQL Server 5.6\bin\mysqldump.exe" -ubackup -pbackup openremdatabasemysql > F:\OpenREMdatabase\backup\%mydate%_%mytime%_openremdatabasemysql.sql

"C:\Program Files\7-Zip\7z.exe" a F:\OpenREMdatabase\backup\%mydate%_%mytime%_openremdatabasemysql.zip F:\OpenREMdatabase\backup\%mydate%_%mytime%_openremdatabasemysql.sql

del F:\OpenREMdatabase\backup\%mydate%_%mytime%_openremdatabasemysql.sql

copy F:\OpenREMdatabase\backup\%mydate%_%mytime%_openremdatabasemysql.zip "\\Srv-mps-001\xls_protect\PATDOSE\OpenREMbackup\"

Headline changes¶

• System

  • Django upgraded to version 1.8
  • Median function added to the database if using PostgreSQL
  • New user-defined display name for each unique system so that rooms with the same DICOM station name are displayed separately
  • Patient name and ID can optionally be stored in system, available for searching and export, but not displayed
  • Patient name, ID and accession number can be stored as a one-way hash, and remain searchable
  • Permission system has become more granular
  • System can now accept non-ASCII characters in protocol names etc
  • Menus have been tidied up
  • Settings file has been updated

• Charts and interface

  • Bar chart data points sorted by frequency, value or name in ascending or descending order
  • CT chart of DLP per requested procedure type
  • CT chart of requested procedure frequency
  • CT chart of CTDIvol per study description
  • Chart data returned using AJAX to make pages more responsive
  • Chart plotting options available via Config menu
  • Charts can now be made full-screen
  • CTDIw phantom size is displayed with the CTDIvol measurement on the CT study detail page
  • Charts show a series called “Blank” when the series name is None
  • Queries for chart data now faster in most situations
  • Histograms can be disabled or enabled for bar charts
  • User-specified number of histogram bins from 2 to 40
  • Mammography chart of average glandular dose vs. compressed thickness
  • Mammography chart showing the number of studies carried out per weekday
  • Fluoroscopy chart of average DAP for each study description
  • Fluoroscopy chart of the frequency of each study description
  • Fluoroscopy chart showing the number of studies carried out per weekday
  • Context specific documentation has been added to the Docs menu

• DICOM Networking

  • Query retrieve function is now built in to query PACS systems or modalities via the Import menu
  • Configuring and running DICOM Store SCP is available and managed in the web interface, but not recommended
  • Documentation improved

• Imports

  • Mammography RDSRs import correctly
  • Mammography imports from images now create an accumulated AGD value per breast
  • GE Senographe DS compression now recorded correctly in Newtons for new imports
  • Philips Allura fluoroscopy RDSRs import correctly, including calculating the exposure time
  • Bi-plane fluoroscopy imports can now be displayed in the web interface
  • Patient height imports from csv now convert from cm to m - previously height was assumed to be cm and inserted into database without change. Existing height data will remain as cm value for csv imports, and m value for RDSR imports
  • Better handling of non-ASCII characters
  • Tube current is now extracted from Siemens Intevo RDSRs

• Exports

  • Patient sex is included in all exports
  • Filters generated by navigating through charts can now be used to filter export data
  • Study description and laterality are now included in mammography exports
  • Bi-fluoroscopy studies can be exported

• Skin dose maps

  • Skin dose maps have been withdrawn from OpenREM version 0.7.0 due to incorrect orientation calculations that need to be fixed before openSkin can be reimplemented into OpenREM

Upgrading an OpenREM server with no internet access¶

mkdir openremfiles
pip install -d openremfiles openrem==0.7.1

pip install --no-index --find-links=openremfiles openrem==0.7.1

Upgrading from version 0.6.0¶

• Back up your database

  • For PostgreSQL you can refer to Backup the database
  • For a non-production SQLite3 database, simply make a copy of the database file

• Stop any Celery workers

• The 0.7.0 upgrade must be made from a 0.6.0 (or later) database, and a schema migration is required:

pip install openrem==0.7.1

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

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

Delete all numbered migration files in openrem’s remapp/migrations folder, leaving the 0002 files ending in .inactive

• If there is no file named __init__.py in the remapp/migrations folder, please create it.
• If you have accidentally deleted the 0002 files ending in .inactive, you can get a new copy from the bitbucket repository.

python manage.py migrate --fake-initial
python manage.py makemigrations remapp
python manage.py migrate remapp --fake

Now rename the file

remapp/migrations/0002_upgrade_0_7_from_0_6.py.inactive

to:

remapp/migrations/0002_upgrade_0_7_from_0_6.py

and then run

python manage.py migrate remapp

• Review the new openremproject/local_settings.py.example file and copy accross the logging section. Then see Log file settings in the install docs.

  If you are using PuTTY on Windows to interact with a linux server, you can select the logging configuration section of the example file with your mouse, and it will be automatically copied to the clipboard. Then open the existing local_settings.py file with nano, move the curser down to the bottom and click the right mouse button to paste.

Upgrading from version 0.7.0 beta 7 or later¶

• Stop any Celery workers
• You will need to do a database migration.

pip install openrem==0.7.1

From the openrem folder (see above):

python manage.py makemigrations remapp
python manage.py migrate remapp

• Review the new local_settings.py.example file and copy accross the logging section. Then see Log file settings in the install docs.

Test web server¶

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

• Ubuntu linux: /usr/local/lib/python2.7/dist-packages/openrem/
• Other linux: /usr/lib/python2.7/site-packages/openrem/
• Linux virtualenv: lib/python2.7/site-packages/openrem/ (remember to activate the virtualenv)
• Windows: C:\Python27\Lib\site-packages\openrem\
• Windows virtualenv: Lib\site-packages\openrem\ (remember to activate the virtualenv)

python manage.py runserver --insecure

python manage.py runserver --insecure 192.168.1.10:8000

ipconfig

ip add

Celery task queue¶

Celery will have been automatically installed with OpenREM, and along with RabbitMQ allows for asynchronous task processing for imports, exports and DICOM networking tasks.

You can put the folder wherever you like, for example you might like to create a /var/log/openrem/ folder on a linux system.

If you are using the built-in Test web server then Celery and the webserver will be running as your user. If you are running a production webserver, such as Apache or nginx on linux, then the user that runs those daemons will need to be able to write to the MEDIA_ROOT and the Celery log files folder. In this case, you need to change the ownership of the folders and change to the right user before running Celery. On Ubuntu:

mkdir /path/to/media/celery  # change as appropriate
sudo chown www-data /path/to/media  # change as appropriate
sudo su -p www-data

Now start celery...

Move into the openrem folder:

• Ubuntu linux: /usr/local/lib/python2.7/dist-packages/openrem/
• Other linux: /usr/lib/python2.7/site-packages/openrem/
• Linux virtualenv: lib/python2.7/site-packages/openrem/ (remember to activate the virtualenv)
• Windows: C:\Python27\Lib\site-packages\openrem\
• Windows virtualenv: Lib\site-packages\openrem\ (remember to activate the virtualenv)

Linux - \ is the line continuation character:

celery multi start default -A openremproject -c 4 -Q default \
--pidfile=/path/to/media/celery/%N.pid --logfile=/path/to/media/celery/%N.log

Windows - celery multi doesn’t work on Windows, and ^ is the continuation character:

celery worker -n default -A openremproject -c 4 -Q default ^
--pidfile=C:\path\to\media\celery\default.pid --logfile=C:\path\to\media\celery\default.log

For production use, see Daemonising Celery below

Set the number of workers (concurrency, -c) as you see fit. The more you have, the more processes (imports, exports, query-retrieve operations etc) can take place simultaneously. However, each extra worker uses extra memory and if you have too many they will be competing for CPU resources too.

To stop the celery queues in Linux:

celery multi stop stores default --pidfile=/path/to/media/celery/%N.pid

For Windows, just press Ctrl+c

You will need to do this twice if there are running tasks you wish to kill.

Celery periodic tasks: beat¶

Celery beat is a scheduler. If it is running, then every 60 seconds a task is run to check if any of the DICOM Store SCP nodes are set to keep_alive, and if they are, it tries to verify they are running with a DICOM echo. If this is not successful, then the Store SCP is started.

To run celery beat, open a new shell and move into the openrem folder:

• Ubuntu linux: /usr/local/lib/python2.7/dist-packages/openrem/
• Other linux: /usr/lib/python2.7/site-packages/openrem/
• Linux virtualenv: lib/python2.7/site-packages/openrem/ (remember to activate the virtualenv)
• Windows: C:\Python27\Lib\site-packages\openrem\
• Windows virtualenv: Lib\site-packages\openrem\ (remember to activate the virtualenv)

Linux:

celery -A openremproject beat -s /path/to/media/celery/celerybeat-schedule \
-f /path/to/media/celery/celerybeat.log \
--pidfile=/path/to/media/celery/celerybeat.pid

Windows:

celery -A openremproject beat -s C:\path\to\media\celery\celerybeat-schedule ^
-f C:\path\to\media\celery\celerybeat.log ^
--pidfile=C:\path\to\media\celery\celerybeat.pid

For production use, see Daemonising Celery below

As with starting the Celery workers, the folder that the pid, log and for beat, schedule files are to be written must already exist and the user starting Celery beat must be able write to that folder.

To stop Celery beat, just press Ctrl+c

Configure the settings¶

• Follow the link presented on the front page to get to the user and group administration.

• After the first users are configured, this link will no longer be presented and instead you can go to Config -> Manage users.

• You will need the superuser username and password you created just after creating the database. The groups are

  • viewgroup can browse the data only
  • importsizegroup can use the csv import facility to add patient height and weight information
  • importqrgroup can use the DICOM query-retrieve facility to pull in studies, as long as they are pre-configured
  • exportgroup can view and export data to a spreadsheet
  • pidgroup can search using patient names and IDs depending on settings, and export with patient names and IDs if they are also a member of the exportgroup
  • admingroup can delete studies, configure DICOM Store/QR settings, configure DICOM keep or delete settings, configure patient ID settings, and abort and delete patient size import jobs. Members of the admingroup no longer inherit the other groups permissions.

• In addition to adding users to these groups, you may like to grant a second user superuser and staff status so that there are at least two people who can manage the users
• Return to the OpenREM interface (click on View site at the top right)

• Go to Config -> DICOM object delete settings and configure appropriately (see Delete objects configuration)
• Go to Config -> Patient ID settings and configure appropriately (see Patient identifiable data)
• If you want to use OpenREM as a DICOM store, or to use OpenREM to query remote systems, go to Config -> Dicom network configuration. For more information go to DICOM Store and QR (not yet up to date)
• With data in the system, you will want to go to Config -> View and edit display names and customise the display names. An established system will have several entries for each device, from each time the software version, station name or other elements changes. See Viewing and editing individual x-ray system display names using the web interface for more information

Start using it!¶

Add some data!

openrem_rdsr.py rdsrfile.dcm

Further instructions¶

Importing dose related data from DICOM files¶

If you are using linux, or for Windows if you have put C:\Python27\;C:\Python27\Lib\site-packages;C:\Python27\Scripts onto your system path, you should be able to import from the command line:

openrem_rdsr.py filename.dcm

openrem_rdsr.py *.dcm

openrem_mg.py filename.dcm

openrem_dx.py filename.dcm

openrem_ctphilips.py filename.dcm

Configure what is stored¶

On the Config menu, select Patient ID settings. The initial settings are as follows:

The default for patient name, ID and date of birth is to not store them. There isn’t an option currently to not store the accession number, though OpenREM continues to work if it is missing.

To store patient identifiable data from now on, select the relevant box and press Submit. If you change the setting again later, then data already stored will remain in the database.

Store encrypted data only¶

If you wish to have the patient name and/or ID available for finding studies relating to a specific patient, but do not need to identify who that patient is, then it is possible to create an ‘encrypted’ version of the ID or name. In this case, a one-way SHA 256 hash is generated and the hash value is stored instead.

If exactly the same name or ID (including spelling, spacing, case etc) occurs more than once, then the same hash will be generated.

The same applies to accession numbers if the option to encrypt the accession number is selected.

Using patient identifiable data¶

Querying for patient studies¶

In the modality pages of the OpenREM web interface, if you are in the pidgroup you will have a filter for patient name and patient ID available:

If the values in the database are not encrypted, then partial search terms can be used as a case-insensitive ‘contains’ query will be applied.

If the values are encrypted, then only the entire string, with exactly the same case, spacing and punctuation will match. This is more likely to be successful with patient ID than with patient name.

Study export with patient identifiers¶

Users in the pidgroup will have extra export buttons available in the modality pages:

If the IDs or names are encrypted, then these columns will contain the hash rather than the original values. However, it will be possible to see if more than one study belongs to one patient as the values should be the same for both. Due to the nature of the algorithm however, a single change in the name or ID - such as an upper case letter instead of a lower case one - will be recorded as a completely different hash value.

Any exports with either patient name or patient ID included will also have a date of birth column.

Navigating the OpenREM web interface¶

Depending on your web server setup, your web interface to OpenREM will usually be at http://yourserver/openrem or if you are using the test web server then it might be at http://localhost:8000/openrem.

The home page for OpenREM should look something like this when it is populated with studies:

By selecting the links in the navigation bar at the top, you can view all of the CT, fluoroscopy or mammography studies. Alternatively, if you click on the station name link (in blue) you can filter to just that source modality.

New in 0.4.0: If you are not logged in, clicking any of the links will bring up the log in page.

Filtering for specific studies¶

This image shows the CT studies view, available to any logged in user, filtered by entering terms in the boxes on the right hand side to show just the studies where the modality model name includes the term ‘soma’:

The search fields can all be used on their own or together, and they are all case insensitive ‘contains’ searches. The exception is the date field, where both from and to have to be filled in (if either are), and the format must be yyyy-mm-dd. There currently isn’t any more complex filtering available, but it does exist as issue 17 for a future release.

The last box below the filtering search boxes is the ordering preference.

Viewing study details¶

By clicking on the study description link (in blue), you can see more details for an individual study:

Not all the details stored for any one study are displayed, just those thought to be most useful. If there are others you’d like to see, add an issue to the tracker.

The final field in the summary at the top is called ‘Test patient indicators?’ When studies are imported the ID and patient name fields are both ignored, but they are parsed to check if they have ‘phy’, ‘test’ or ‘qa’ in them to help exclude them from the data analysis. If they do, then this information is added to the field and is displayed both in the web interface as a Test patient indicator and in the Excel export. The name and ID themselves are not reproduced, simply the presence of one of the key words. Therefore a patient named ‘Phyliss’ would trigger this, but only ‘Phy’ would be reproduced in this field. Other fields will also help to confirm whether a study is for a real patient such as the lack of an Accession Number and an unusual patient age.

Chart types¶

1. Bar chart of average values across a number of categories¶

Figure 1: Bar chart of mean DLP per acquisition

An example of mean DAP per acquisition type is shown in figure 1.

Figure 2: Bar chart sorting options

Below each bar chart there are options to sort the order of the data. This can be ascending or descending by average value, size of data sample, or alphabetically (figure 2).

Clicking on an entry in the bar chart legend toggles the display of the corresponding series on the chart.

Figure 3: Histogram of abdomen DLP values

Clicking on an individual data point on a bar chart will take you to a histogram of the data for that point so that you can see the shape of the value’s distribution (figure 3).

Bar charts can be plotted with a series per x-ray system (figure 4). This can be toggled using the Plot a series per system checkbox in the Chart options.

Clicking the left-hand mouse button on the chart background and dragging left or right selects part of the series. Releasing the mouse button zooms in on this selection. A Reset zoom button appears when zoomed in: clicking this resets the chart so that the full series can be seen again. The zoom feature works on both the main series and the histograms. The zooming can be useful when there is a category on the chart that has a very low value compared to others. Zooming in on this category will enable the low values to be seen, as the chart rescales the y-axis after the zoom.

Figure 4: Bar chart of mean DLP (one system per series)

If the the bar chart that you are viewing shows more than one series then clicking on a category name on the x-axis will take you to a plot that shows multiple histograms: one for each series (figure 5).

Figure 5: Histogram of abdomen DLP values, one series per system

The histogram data can be plotted as absolute values, or be normalised to a value of 1.0 (figure 6). This can be toggled by clicking on the button that is shown below the histogram plots. The normalisation can be useful when trying to compare the shape of several histograms, especially when some histograms have much less data than others.

Figure 6: Normalised histogram of abdomen DLP, one series per system

Each histogram data point includes a text link that appears when the mouse pointer moves over it. Clicking on this link will filter the displayed studies, showing those that correspond to what is contained in the histogram bin.

Clicking on a legend entry toggles the visibility of the corresponding series.

2. Pie chart showing the frequency of each item in a category¶

Figure 7: Pie chart of acquisition frequency

Figure 7 shows a pie chart of the number of acquisitions made for every acquisition protocol present in the tabulated data.

Clicking on any of the pie chart segments will filter the displayed studies, showing only the studies that correspond to what is contained in that segment. As for the bar charts, this doesn’t work perfectly, as the category filtering isn’t exact.

3. Line chart showing how an average value changes over time¶

Figure 8: Line chart of mean DLP per study type over time

A line is plotted for each category, with a point calculated every day, week, month or year. This can be a good way of looking at how things have changed over time. For example, the mean DLP of each study type, calculated with a data point per month is shown in figure 8.

Clicking the left-hand mouse button on the chart and dragging left or right across a range of dates and then releasing the mouse button will zoom in on that selection.

Clicking on a legend entry toggles the visibility of the corresponding series.

4. Pie chart showing the number of events per day of the week¶

Figure 9: Pie chart of study workload per day of the week

Figure 10: Pie chart of study workload per hour in a day

Each segment represents a day of the week, and shows the number of events that have taken place on that day (figure 9). Clicking on one of the segments will take you to a pie chart that shows the number of events per on that day (figure 10).

5. Scatter plot showing one value vs. another¶

Figure 11: Scatter plot of average glandular dose vs. compressed thickness

Figure 12: Scatter plot of average glandular dose vs. compressed thickness; one series per system

This plot type shows a data point per event (figure 11). The series name and data values are shown when the mouse cursor is positioned over a data point.

These can be plotted with a series per x-ray system (figure 12). This can be toggled using the Plot a series per system checkbox in the Chart options.

Clicking the left-hand mouse button on the chart and dragging a rectangular region will zoom in on that selection of the chart. A Reset zoom button appears when zoomed in: clicking this resets the chart so that the full series can be seen again.

Clicking on a system’s legend entry toggles the display of the corresponding series on the chart.

Exporting chart data¶

An image file of a chart can be saved using the menu in the top-right hand side of any of the charts. The same menu can be used to save the data used to plot a chart: the data can be downloaded in either csv or xls format.

New in 0.7.0¶

• Charts for fluoroscopy and mammography.
• New scatter plot chart type.
• Chart plotting options can be configured by choosing Chart options from the User options menu at the top of the OpenREM homepage.
• Chart average values can be set to either mean or median. Bar charts can be configured to plot both mean and median values as seperate series.
• Bar charts can be plotted with a series per x-ray system. This option can be switched on or off via the Chart options.
• The number of histogram data bins can be set to a value between 2 and 40 in Chart options. The default value is 20.
• Histogram calculation can be switched on or off in Chart options. The default is off. Performance is significantly better when set to off.
• Histogram plots can toggle between absolute or normalised values via the Toggle Normalised histograms button that is visible when viewing a histogram plot.
• The data in the bar charts can be sorted interactively by clicking on the sorting options below the individual chart. The default sorting type and direction can be set by choosing the Chart options item from the User options menu on the OpenREM homepage.
• Individual charts can be displayed full-screen by clicking on the Toggle fullscreen button that is positioned below each chart.
• The chart plotting status is displayed on the OpenREM homepage.
• The colours used for plotting have been updated.

Chart options¶

Figure 13: OpenREM chart options

Chart options can be configured by choosing the Chart options item from the User options menu on the OpenREM homepage (figure 13).

CT and radiographic plot options can also be set from their respective summary pages.

The first option, Plot charts?, determines whether any plots are shown. This also controls whether the data for the plots is calculated by OpenREM. Some plot data is slow to calculate when there is a large amount of data: some users may prefer to leave Plot charts? off for performance reasons. Plot charts? can be switched on and activated with a click of the Submit button after the data has been filtered.

Figure 14: Switching charts off

The user can also switch off chart plotting by clicking on the Switch charts off link in the User options menu in the navigation bar at the top of any OpenREM page, as shown in figure 14.

The user can choose whether the data displayed on the charts is the mean, median or both by using the drop-down Average to use selection. Only the bar charts can display both mean and median together. Other charts display just median data when this option is selected.

The charts can be sorted by either bar height, frequency or alphabetically by category. The default sorting direction can be set to ascending or descending using the drop-down list near the top of the chart options.

A user’s chart options can also be configured by an administrator via OpenREM’s user administration page.

Chart types - CT¶

• Bar chart of average DLP for each acquisition protocol (all systems combined)
• Bar chart of average DLP for each acquisition protocol (one series per system)
• Pie chart of the frequency of each acquisition protocol
• Pie chart showing the number of studies carried on each day of the week
• Line chart showing the average DLP of each study name over time
• Bar chart of average CTDI_vol for each acquisition protocol
• Bar chart of average DLP for each study name
• Pie chart of the frequency of each study name
• Bar chart of average DLP for each requested procedure
• Pie chart of the frequency of each requested procedure

Chart types - radiography¶

• Bar chart of average DAP for each acquisition protocol
• Pie chart of the frequency of each acquisition protocol
• Bar chart of average DAP for each study description
• Pie chart of the frequency of each study description
• Bar chart of average DAP for each requested procedure
• Pie chart of the frequency of each requested procedure
• Bar chart of average kVp for each acquisition protocol
• Bar chart of average mAs for each acquisition protocol
• Pie chart showing the number of studies carried out per weekday
• Line chart of average DAP of each acquisition protocol over time
• Line chart of average mAs of each acquisition protocol over time
• Line chart of average kVp of each acquisition protocol over time

Chart types - fluoroscopy¶

• Bar chart of average DAP for each study description
• Pie chart of the frequency of each study description
• Pie chart showing the number of studies carried out per weekday

Chart types - mammography¶

• Scatter plot of average glandular dose vs. compressed thickness for each acquisition
• Pie chart showing the number of studies carried out per weekday

Performance notes¶

Functionality that will be available¶

• Skin dose map data calculated to the surface of a simple geometric phantom using the in-built openSkin routines (3D phantom)
• Phantom dimensions calculated from the height and mass of the patient
• Data can be calculated on import to OpenREM, or on demand when a study is viewed
• 3D skin dose map data shown graphically as a 2D image and a 3D model
• The user can change the maximum and minimum displayed dose; alternatively, window level and width can be adjusted
• A colour dose scale is shown with a selection of colour schemes
• The skin dose map section can be displayed full-screen
• The calculated peak skin dose, phantom dimensions and patient height and mass used for the calculations are shown in the top left hand corner of the skin dose map
• If skin dose map display is disabled then fluoroscopy study data can be exported in a format suitable for the stand-alone openSkin routines

The phantom consists of a cuboid with one semi-cylinder on each side (see 3D phantom section of phantom design on the openSkin website for details). A default height of 1.786 m and mass of 73.2 kg are used if patient-specific data are unavailable.

2D visualisation of the 3D data¶

This is a 2D view of the whole surface of the 3D phantom, as though the phantom surface has been peeled off and laid out flat. The 2D visualisation includes the following features:

• The skin dose at the mouse pointer is shown as a tool-tip
• Moving the mouse whilst holding down the left-hand mouse button changes the window level and width of the displayed skin dose map
• An overlay indicating the phantom regions and orientation can be toggled on and off. This indicates the phantom anterior, left, posterior and right sides, and also shows the superior and inferior ends
• The current view can be saved as a png file

Figure 1: 2D visualisation of the 3D data

Figure 2: Phantom region overlay

3D visualisation¶

This is a 3D view of the phantom that was used for the calculations, with the skin dose map overlaid onto the surface. The 3D visualisation includes the following features:

• Moving the mouse whilst holding down the left-hand mouse button rotates the 3D model
• Using the mouse wheel zooms in and out
• A simple 3D model of a person is displayed in the bottom left corner. This is to enable the viewer to orientate themselves when viewing the 3D skin dose map
• The current view can be saved as a png file

Figure 3: 3D visualisation of the data

Figure 4: Colour scale choices

Skin dose map settings¶

There are two skin dose map options that can be set by an OpenREM administrator via the Skin dose map settings option in the Config menu:

• Enable skin dose maps
• Calculate skin dose maps on import

The first of these sets whether skin dose map data is calculated, and also switches the display of skin dose maps on or off. The second option controls whether the skin dose map data is calculated at the point when a new study is imported into OpenREM.

When skin dose maps are enabled:

• When a user views the details of a fluoroscopy study OpenREM looks for a skin dose map pickle file on the OpenREM server in the skin_maps subfolder of MEDIA_ROOT that corresponds to the study being viewed. If found, the skin dose map data in the pickle file is loaded and displayed. The skin_maps folder is created if it does not exist
• If a pickle file is not found then OpenREM calculates skin dose map data. These calculations can take some time. They are carried out in the background: an animated graphic is shown during the calculations. On successful calculation of the data the skin dose map is displayed. A pickle file containing the data is saved in the server’s skin_maps subfolder of MEDIA_ROOT. The file name is of the form skin_map_XXXX.p, where XXXX is the database primary key of the study
• For subsequent views of the same study the data in the pickle file is loaded, rather than re-calculating the data, making the display of the skin dose map much quicker

When calculation on import is enabled:

• OpenREM calculates the skin dose map data for a fluoroscopy study as soon as it arrives in the system
• A pickle file containing the data is saved in the skin_maps subfolder of MEDIA_ROOT
• Users viewing the details of a study won’t have to wait for the skin dose map data to be calculated

Exporting data to openSkin¶

If skin dose maps are disabled the user you are presented with the option of exporting the study data as a csv file for use with a stand-alone installation of openSkin. Select the fluoroscopy study you wish to create the exposure incidence map for and go to the detail view. Then click on the link to create the OpenSkin export (figure 5).

Figure 5: Export from OpenREM to openSkin

Instructions for openSkin¶

Download the latest version as a zip file from openSkin downloads. At the time of release for OpenREM 0.7.0 the current openSkin beta was dated 14th September 2016. The application referred to here will only work on Windows.

• Extract the contents of the zip file into a folder on your computer and run the openSkin.exe executable
• Choose a phantom type: 3D or flat. See phantom design for details
• Select the source csv file - this should be the one exported from OpenREM
• Select the output folder - this should already exist as it can’t be created in the dialogue
• Wait! Depending on the number of events in the export and the power of your machine, this can take a few minutes

Two files will be produced - a textfile called skin_dose_results.txt and a small image called skin_dose_map.png

File created    : 04/21/15 17:42:45
Data file       : C:/Users/[...]/exports-2015-04-21-OpenSkinExport20150421-162805246134.csv
Phantom         : 90.0x70.0 3d phantom
Peak dose (Gy)  :                  0.50844405521
Cells > 3 Gy    :                              0
Cells > 5 Gy    :                              0
Cells > 10 Gy   :                              0

Incidence map image file¶

The image file will be a small 70x90 px PNG image if you used the 3D phantom, or 150 x 50 px PNG if you used the 2D phantom. With both, the head end of the table is on the left.

The image is scaled so that black is 0 Gy and white is 10 Gy. For most studies, this results in an incidence map that is largely black. However, if you use GIMP or ImageJ or similar to increase the contrast, you will find that the required map is there.

A native and ‘colour equalised’ version of the same export are shown below:

Limitations¶

Skin dose map calculations do not currently work for all systems. Siemens Artis Zee data is known to work. If skin dose maps do not work for your systems then please let us know via the OpenREM Google Group.

openSkin is yet to be validated independently - if this is something you want to do, please do go ahead and feed back your findings to Jonathan Cole at jacole.

Exporting to csv and xlsx sheets¶

If you are logged in as a user in the exportgroup or the admingroup, the export links will be available near the top of the modality filter pages in the OpenREM interface. The following exports are currently available (version 0.5.0)

• CT basic, single sheet csv
• CT advanced, XLSX muliple-sheets
• Fluoroscopy basic, single sheet csv
• Mammography, single sheet csv
• Mammography NHSBSP, single sheet csv designed to satisfy NHSPSB reporting
• Radiographic, single sheet csv
• Radiographic, XLSX multiple sheets

For CT and radiographic, the XLSX export has multiple sheets. The first sheet contains a summary of all the study descriptions, requested procedures and series protocol names contained in the export:

This information is useful for seeing what data is in the spreadsheet, and can also be used to prioritise which studies or protocols to analyse based on frequency.

The second sheet of the exported file lists all the studies, with each study taking one line and each series in the study displayed in the columns to the right.

The remainder of the file has one sheet per series protocol name. Each series is listed one per line. If a single study has more than one series with the same protocol name, then the same study will appear on more than one line.

Clicking the link for an export redirects you to the Exports page, which you can also get to using the link at the top right of the navigation bar:

Whilst an export is being processed, it will be listed in the first table at the top. The current status is displayed to indicate export progress. If an export gets stuck for whatever reason, you may be able to abort the process by clicking the ‘Abort’ button. However this does not always cause an active export to terminate - you may find it completes anyway!

Completed exports are then listed in the second table, with a link to download the csv or xlsx file.

When the export is no longer needed, it can be deleted from the server by ticking the delete checkbox and clicking the delete button at the bottom:

Server 500 errors¶

Turn on debug mode

Locate and edit your local_settings file

• Ubuntu linux: /usr/local/lib/python2.7/dist-packages/openrem/openremproject/local_settings.py

• Other linux: /usr/lib/python2.7/site-packages/openrem/openremproject/local_settings.py

• Linux virtualenv: lib/python2.7/site-packages/openrem/openremproject/local_settings.py

• Windows: C:\Python27\Lib\site-packages\openrem\openremproject\local_settings.py

• Windows virtualenv: Lib\site-packages\openrem\openremproject\local_settings.py

• Change the line:

  # DEBUG = True
  

• to:

  DEBUG = True
  

This will render a debug report in the browser - usually revealing the problem.

Once the problem is fixed, change DEBUG to False, or comment it again using a #. If you leave debug mode in place, the system is likely to run out of memory.

Query-retrieve issues¶

Refer to the Troubleshooting: openrem_qr.log documentation

OpenREM DICOM storage nodes¶

Refer to the Troubleshooting: openrem_store.log documentation

Log files¶

Log file location, naming and verbosity were configured in the local_settings.py configuration - see the Log file configuration docs for details.

If the defaults have not been modified, then there will be three log files in your MEDIAROOT folder which you configured at installation. See the install config section on Location for imports and exports for details.

The openrem.log has general logging information, the other two are specific to the DICOM store and DICOM query-retrieve functions if you are making use of them.

Indices and tables¶

• Index
• Module Index
• Search Page

Version history change log¶

OpenREM version history¶

0.7.1 (2016-06-10)¶

• #403 Now deals with PersonName fields with latin-1 extended characters correctly
• #402 Skin dose map data pickle files saved using gzip compression to save space
• #401 Updated skin dose map documentation to say it won’t be in this release
• #400 Strings are encoded as UTF-8 before being hashed to prevent errors with non-ASCII characters
• #399 Migration file brought up to date for 0.6 to 0.7 upgrades
• #398 Skin exposure maps are now stored in folders (feature postponed for future release)
• #397 Skin exposure maps no longer available until orientation errors are fixed
• #396 Charts: zooming on bar charts of average value vs. category now works
• #395 Docs: offline Windows install instructions created, plus offline upgrade instructions
• #394 Charts: made charts resize to fit containing div when browser is resized
• #392 Charts: normalised histogram tooltip now correctly reports frequency
• #391 Basic troubleshooting is now documented
• #390 Charts: mammography and fluoroscopy charts added
• #389 Charts: series without a name are now plotted under the name of Blank rather than not being plotted at all
• #387 Added laterality to mammography exports
• #385 Fixed issue with non-ASCII letters in RDSR sequence TextValue fields
• #384 Fluoro exports for OpenSkin only consider copper filters now
• #383 Refreshed settings.py to django 1.8 including updating template settings and TEMPLATE_CONTEXT_PROCESSORS
• #380 Tube current now extracted from Siemens Intevo RDSR despite non-conformance
• #379 Exposure time now populated for fluoro if not supplied by RDSR
• #378 The display name of multiple systems can now be updated together using a single new name
• #376 Corrected an ill-advised model change
• #374 CTDIw phantom size now displayed in CT detail view
• #373 Charts in some releases used GT rather than greater than or equal to for start date, now fixed
• #372 Mammography studies now record an accumulated AGD per breast. Existing joint accumulated AGD values won’t be changed. Ordering by Accumulated AGD now creates an entry per accumulated AGD, one per breast
• #371 Mammo RDSR generates average mA where not recorded, mammo image populates mA
• #370 Added study description to mammography export
• #369 Bi-plane fluoroscopy studies now export correctly
• #368 Mammo RDSR now imports correctly
• #365 Tube filtration is now displayed in the RF detail view
• #364 Philips Allura fluorscopy RDSRs now import correctly
• #362 Display of RF where bi-plane RDSRs have been imported no longer crash the interface
• #360 Charts: saving data from average data charts as csv or xls now includes frequency values
• #359 Added missing ‘y’ to query retrieve command line help
• #358 Charts: chart sorting links and instructions now hidden when viewing histograms
• #357 Charts: button to return from histogram now displays the name of the main chart
• #356 Charts: histogram normalise button appears for all appropriate charts
• #355 Charts: sorting now works as expected for plots with a series per system
• #352 Fixed CT xlsx exports that had complete study data in each series protocol sheet (from earlier beta)
• #351 Charts: simplified chart JavaScript and Python code
• #350 DICOM networking documented for use with 3rd party store and advanced use with native
• #348 Study delete confirmation page now displays total DAP for DX or CR radiographic studies
• #346 Charts: exporting a chart as an image no longer requires an internet connection
• #345 CSV size imports in cm are now stored as m in the database. Interface display of size corrected.
• #343 Charts: user can now specify number of histogram bins in the range of 2 to 40
• #342 Charts: improved the colours used for plotting chart data
• #340 Fixed store failure to save due to illegal values in Philips private tags, improved exception code
• #339 Improved extraction of requested procedure information for radiographic studies
• #338 Fix Kodak illegally using comma in filter thickness values
• #335 DICOM Store keep_alive and echo_scu functions now log correctly
• #334 Fixed issue with tasks needing to be explicitly named
• #333 Fixed StoreSCP not starting in beta 11 error
• #332 Charts: some charts can now be plotted with a series per x-ray system
• #331 Keep_alive tasks are now discarded if not executed, so don’t pile up
• #329 All existing logging is now done via the same log files
• #328 Store SCP no longer uses Celery tasks
• #327 Celery workers now only take one task at a time
• #325 Charts: switching charts off now leaves the user on the same page, rather than going to the home page
• #324 Charts: forced chart tooltip background to be opaque to make reading the text easier
• #320 The week now begins on Monday rather than Sunday on date form fields
• #316 Query retrieve function can now exclude and include based on strings entered
• #315 Charts: made size of exported chart graphics follow the browser window size
• #314 One version number declaration now used for distribute, docs and interface
• #313 Replaced non-working function with code to extract SeriesDescription etc in query response message
• #312 Display names are now grouped by modality
• #311 Queries are deleted from database after a successful C-Move
• #310 Series level QR feedback now presented. Any further would require improvements in pynetdicom
• #309 StoreSCP now deals safely with incoming files with additional transfer syntax tag
• #308 Secondary capture images that don’t have the manufacturer field no longer crash the StoreSCP function
• #306 Charts: added a button to each chart to toggle full-screen display
• #305 Added links to documentation throughout the web interface
• #304 Date of birth is now included in all exports that have either patient name or ID included
• #303 Fixed a typo in 0.6.0 documents relating to the storescp command
• #302 Improved handling of Philips Dose Info objects when series information sequence has UN value representation
• #301 Charts: fixed bug that could stop average kVp and mAs radiographic plots from working
• #300 Calling AE Title for Query Retrieve SCU is now configured not hardcoded
• #299 Hash of MultiValued DICOM elements now works
• #298 Added ordering by accumulated AGD for mammographic studies
• #297 Fixed ordering by Total DAP for radiographic studies
• #296 StoreSCP now logs an error message and continues if incoming file has problems
• #295 Charts: fixed bug that arose on non-PostgreSQL databases
• #294 Harmonised time display between filter list and detail view, both to HH:mm
• #292 Added keep-alive and auto-start to DICOM stores
• #291 Charts: fixed issue with CTDI and DLP not showing correct drilldown data
• #290 Added new tables and fields to migration file, uses #288 and median code from #241
• #289 Crispy forms added into the requires file
• #288 Added device name hashes to migration file
• #286 Increased granularity of permission groups
• #285 Tidied up Options and Admin menus
• #284 Fixed DICOM Query that looped if SCP respected ModalitiesInStudy
• #282 Missing javascript file required for IE8 and below added
• #281 Added check to import function to prevent extract failure
• #280 Fixed typo in mammography export
• #279 Charts: Fixed issue with median CTDI series from appearing
• #278 Charts: Fixed javascript namespace pollution that caused links to fail
• #277 Overhaul of acquisition level filters to get tooltip generated filters to follow through to export
• #276 Unique fields cannot have unlimited length in MySQL - replaced with hash
• #274 Charts: Fixed legend display issue
• #273 Charts: Added plots of average kVp and mAs over time for DX
• #272 Tweak to display of exam description for DX
• #271 Fixed DX import failure where AcquisitionDate or AcquisitionTime are None
• #270 Django 1.8 Admin site has a ‘view site’ link. Pointed it back to OpenREM
• #268 Improved population of procedure_code_meaning for DX imports
• #266 DICOM C-Store script added back in - largely redundant with web interface
• #265 DICOM Store and Query Retrieve services documented
• #263 Settings for keeping or deleting files once processed moved to database and web interface
• #262 Dealt with issue where two exposures from the same study would race on import
• #260 Fixed issue where import and export jobs would get stuck behind StoreSCP task in queue
• #259 Link to manage users added to Admin menu
• #258 Fixed DX import error where manufacturer or model name was not provided
• #257 Documentation update
• #256 Fixed errors with non-ASCII characters in imports and query-retrieve
• #255 Charts: Small y-axis values on histograms are more visible when viewing full-screen
• #254 Charts: Simplified chart data processing in the templates
• #253 Charts: AJAX used to make pages responsive with large datasets when charts enabled
• #252 Fixed duplicate entries in DX filtered data for studies with multiple exposures
• #248 Charts: can now be ordered by frequency or alphabetically
• #247 Fixed incorrect reference to manufacturer_model_name
• #246 Charts: Added median data for PostgreSQL users
• #245 Fixed error in csv DX export
• #244 Fixed issue where scripts wouldn’t function after upgrade to Django 1.8
• #243 Added distance related data to DX exports
• #242 Distance source to patient now extracted from DX images
• #241 Charts: Median values can be plotted for PostgreSQL users
• #240 Charts: Improved DAP over time calculations
• #239 Configurable equipment names to fix multiple sources with the same station name
• #237 Charts: Tidied up plot data calculations in views.py
• #235 Added patient sex to each of the exports
• #234 Charts: Fixed error with datetime combine
• #232 Charts: on or off displayed on the home page
• #231 Charts: made links from requested procedure frequency plot respect the other filters
• #230 Fixed error in OperatorsName field in DICOM extraction
• #229 Charts: Added chart of DLP per requested procedure
• #223 Charts: speed improvement for weekday charts
• #217 Charts: Further code optimisation to speed up calculation time
• #207 DICOM QR SCU now available from web interface
• #206 DICOM Store SCP configuration now available from web interface
• #183 Added options to store patient name and ID, and options to hash name, ID and accession number
• #171 Root URL now resolves so /openrem is not necessary
• #151 Suspected non-patient studies can now be filtered out
• #135 GE Senographe DS now correctly records compression force in Newtons for new imports
• #120 Improved testing of data existing for exports
• #118 Upgraded to Django 1.8
• #70 User is returned to the filtered view after deleting a study
• #61 Skin dose maps for fluoroscopy systems can now be calculated and displayed

0.6.2 (2016-01-27)¶

• #347 Django-filter v0.12 has minimum Django version of 1.8, fixed OpenREM 0.6.2 to max django-filter 0.11
• #341 Changed references to the OpenSkin repository for 0.6 series.

0.6.1 (2015-10-30)¶

• #303 Corrected name of Store SCP command in docs

0.6.0 (2015-05-14)¶

• #227 Fixed import of RDSRs from Toshiba Cath Labs
• #226 Charts: Updated Highcharts code and partially fixed issues with CTDIvol and DLP combined chart
• #225 Charts: Added link from mAs and kVp histograms to associated data
• #224 Charts: Added link from CTDIvol histograms to associated data
• #221 Charts: Fixed issue where filters at acquisition event level were not adequately restricting the chart data
• #219 Charts: Fixed issue where some charts showed data beyond the current filter
• #217 Charts: Code optimised to speed up calculation time
• #216 Fixed typo that prevented import of RSDR when DICOM store settings not present
• #215 Charts: Fixed x-axis labels for mean dose over time charts
• #214 Charts: Improved consistency of axis labels
• #213 Fixed admin menu not working
• #212 Charts: Created off-switch for charts
• #210 OpenSkin exports documented
• #209 Charts: Fixed server error when CT plots switched off and filter form submitted
• #208 Charts: Fixed blank chart plotting options when clicking on histogram tooltip link
• #205 Charts: Fixed issue of histogram tooltip links to data not working
• #204 Charts: Fixed issue of not being able to export with the charts features added
• #203 Charts: Fixed display of HTML in plots issue
• #202 Charts: Added mean CTDIvol to charts
• #200 Charts: Now exclude Philips Ingenuity SPRs from plots
• #196 Added comments and entrance exposure data to DX export
• #195 Fixed error with no users on fresh install
• #194 Added more robust extraction of series description from DX
• #193 Charts: Fixed reset of filters when moving between pages
• #192 Created RF export for OpenSkin
• #191 Charts: Factored out the javascript from the filtered.html files
• #190 Charts: Added time period configuration to dose over time plots
• #189 Charts: Fixed plotting of mean doses over time when frequency not plotted
• #187 Charts: Merged the charts work into the main develop branch
• #186 Fixed duplicate data in DX exports
• #179 Charts: Added kVp and mAs plots for DX
• #177 Charts: Fixed issue with date ranges for DX mean dose over time charts
• #176 Charts: Added link to filtered dataset from mean dose over time charts
• #175 Charts: Allowed configuration of the time period for mean dose trend charts to improve performance
• #174 Charts: Fixed number of decimal places for mean DLP values
• #173 Charts: Fixed plot of mean DLP over time y-axis issue
• #170 Charts: Added plot of mean dose over time
• #169 Charts: Improved chart colours
• #157 Charts: Added chart showing number of studies per day of the week, then hour in the day
• #156 Charts: Fixed issue with some protocols not being displayed
• #155 Charts: Added chart showing relative frequency of protocols and study types
• #140 Charts: Added configuration options
• #139 Charts: Link to filtered dataset from histogram chart
• #138 Charts: Number of datapoints displayed on tooltip
• #135 Mammography compression force now only divides by 10 if model contains senograph ds Change in behaviour
• #133 Documented installation of NumPy, initially for charts
• #41 Preview of DICOM Store SCP now available
• #20 Modality sections are now suppressed until populated

0.5.1 (2015-03-12)¶

• #184 Documentation for 0.5.1
• #180 Rename all reverse lookups as a result of #62
• #178 Added documentation regarding backing up and restoring PostgreSQL OpenREM databases
• #172 Revert all changes made to database so #62 could take place first
• #165 Extract height and weight from DX, height from RDSR, all if available
• #161 Views and exports now look for accumulated data in the right table after changes in #159 and #160
• #160 Created the data migration to move all the DX accumulated data from TID 10004 to TID 10007
• #159 Modified the DX import to populate TID 10007 rather than TID 10004. RDSR RF already populates both
• #158 Demo website created by DJ Platten: http://demo.openrem.org/openrem
• #154 Various decimal fields are defined with too few decimal places - all have now been extended.
• #153 Changed home page and modality pages to have whole row clickable and highlighted
• #150 DJ Platten has added Conquest configuration information
• #137 Carestream DX multiple filter thickness values in a DS VR now extracted correctly
• #113 Fixed and improved recording of grid information for mammo and DX and RDSR import routines
• #62 Refactored all model names to be less than 39 characters and be in CamelCase to allow database migrations and to come into line with PEP 8 naming conventions for classes.

0.5.0 (2014-11-19)¶

• Pull request from DJ Platten: Improved display of DX data and improved export of DX data
• #132 Fixed mammo export error that slipped in before the first beta
• #130 Only creates ExposureInuAs from Exposure if Exposure exists now
• #128 Updated some non-core documentation that didn’t have the new local_settings.py reference or the new openremproject folder name
• #127 DX IOD studies with image view populated failed to export due to lack of conversion to string
• #126 Documentation created for the radiographic functionality
• #125 Fixes issue where Hologic tomo projection objects were dropped as they have the same event time as the 2D element
• #123 Fixed issue where filters came through on export as lists rather than strings on some installs
• #122 Exports of RF data should now be more useful when exporting to xlsx. Will need refinement in the future
• #26 Extractors created for radiographic DICOM images. Contributed by DJ Platten
• #25 Views and templates added for radiographic exposures - either from RDSRs or from images - see #26. Contributed by DJ Platten
• #9 Import of *.dcm should now be available from Windows and Linux alike

0.4.3 (2014-10-01)¶

• #119 Fixed issue where Celery didn’t work on Windows. Django project folder is now called openremproject instead of openrem
• #117 Added Windows line endings to patient size import logs
• #113 Fixed units spelling error in patient size import logs
• #112 File system errors during imports and exports are now handled properly with tasks listed in error states on the summary pages
• #111 Added abort function to patient size imports and study exports
• #110 Converted exports to use the FileField handling for storage and access, plus modified folder structure.
• #109 Added example MEDIA_ROOT path for Windows to the install docs
• #108 Documented ownership issues between the webserver and Celery
• #107 Documented process for upgrading to 0.4.2 before 0.4.3 for versions 0.3.9 or earlier
• #106 Added the duration of export time to the exports table. Also added template formatting tag to convert seconds to natural time
• #105 Fixed bug in Philips CT import where decimal.Decimal was not imported before being used in the age calculation
• #104 Added documentation for the additional study export functions as a result of using Celery tasks in task #19 as well as documentation for the code
• #103 Added documentation for using the web import of patient size information as well as the new code
• #102 Improved handling of attempts to process patient size files that have been deleted for when users go back in the browser after the process is finished
• #101 Set the security of the new patient size imports to prevent users below admin level from using it
• #100 Logging information for patient size imports was being written to the database - changed to write to file
• #99 Method for importing remapp from scripts and for setting the DJANGO_SETTINGS_MODULE made more robust so that it should work out of the box on Windows, debian derivatives and virtualenvs
• #98 Versions 0.4.0 to 0.4.2 had a settings.py.new file to avoid overwriting settings files on upgrades; renaming this file was missing from the installation documentation for new installs
• #97 Changed the name of the export views file from ajaxviews as ajax wasn’t used in the end
• #96 Changed mammo and fluoro filters to use named fields to avoid needing to use the full database path
• #93 Set the security of the new exports to prevent users below export level from creating or downloading exports
• #92 Add NHSBSP specific mammography csv export from Jonathan Cole - with Celery
• #91 Added documentation for Celery and RabbitMQ
• #90 Added delete function for exports
• #89 Added the Exports navigation item to all templates, limited to export or admin users
• #88 Converted fluoroscopy objects to using the Celery task manager after starting with CT for #19
• #87 Converted mammography objects to using the Celery task manager after starting with CT for #19
• #86 Digital Breast Tomosynthesis systems have a projections object that for Hologic contains required dosimetry information
• #85 Fix for bug introduced in #75 where adaption of ptsize import for procedure import broke ptsize imports
• #74 ‘Time since last study’ is now correct when daylight saving time kicks in
• #39 Debug mode now defaults to False
• #21 Height and weight data can now be imported through forms in the web interface
• #19 Exports are now sent to a task manager instead of locking up the web interface

Reopened issue¶

• #9 Issue tracking import using *.dcm style wildcards reopened as Windows cmd.exe shell doesn’t do wildcard expansion, so this will need to be handled by OpenREM in a future version

0.4.2 (2014-04-15)¶

• #83 Fix for bug introduced in #73 that prevents the import scripts from working.

0.4.1 (2014-04-15)¶

• #82 Added instructions for adding users to the release notes

0.4.0 (2014-04-15)¶

Note

• #64 includes changes to the database schema and needs a user response - see version 0.4.0 release notes
• #65 includes changes to the settings file which require settings information to be copied and files moved/renamed - see version 0.4.0 release notes

• #80 Added docs for installing Apache with auto-start on Windows Server 2012. Contributed by JA Cole
• #79 Updated README.rst instructions
• #78 Moved upgrade documentation into the release notes page
• #77 Removed docs builds from repository
• #76 Fixed crash if exporting from development environment
• #75 Fixed bug where requested procedure wasn’t being captured on one modality
• #73 Made launch scripts and ptsizecsv2db more robust
• #72 Moved the secret key into the local documentation and added instructions to change it to release notes and install instructions
• #71 Added information about configuring users to the install documentation
• #69 Added documentation about the new delete study function
• #68 Now checks sequence code meaning and value exists before assigning them. Thanks to JA Cole
• #67 Added ‘Contributing authors’ section of documentation
• #66 Added ‘Release notes’ section of documentation, incuding this file
• #65 Added new local_settings.py file for database settings and other local settings
• #64 Fixed imports failing due to non-conforming strings that were too long
• #63 The mammography import code stored the date of birth unnecessarily. Also now gets decimal_age from age field if necessary
• #60 Removed extraneous colon from interface data field
• #18 Studies can now be deleted from the web interface with the correct login
• #16 Added user authentication with different levels of access
• #9 Enable import of *.dcm

0.3.9 (2014-03-08)¶

Note

#51 includes changes to the database schema – make sure South is in use before upgrading. See http://docs.openrem.org/page/upgrade.html

• #59 CSS stylesheet referenced particular fonts that are not in the distribution – references removed
• #58 Export to xlsx more robust - limitation of 31 characters for sheet names now enforced
• #57 Modified the docs slightly to include notice to convert to South before upgrading
• #56 Corrected the mammography target and filter options added for issue #44
• #53 Dates can now be selected from a date picker widget for filtering studies
• #52 Split the date field into two so either, both or neither can be specified
• #51 Remove import modifications from issue #28 and #43 now that exports are filtered in a better way after #48 and #49 changes.
• #50 No longer necessary to apply a filter before exporting – docs changed to reflect this
• #49 CSV exports changed to use the same filtering routine introduced for #48 to better handle missing attributes
• #48 New feature – can now filter by patient age. Improved export to xlsx to better handle missing attributes
• #47 Install was failing on pydicom – fixed upstream

0.3.8 (2014-03-05)¶

• – File layout modified to conform to norms
• #46 Updated documentation to reflect limited testing of mammo import on additional modalities
• #45 mam.py was missing the licence header - fixed
• #44 Added Tungsten, Silver and Aluminum to mammo target/filter strings to match – thanks to DJ Platten for strings
• #43 Mammography and Philips CT import and export now more robust for images with missing information such as accession number and collimated field size
• #42 Documentation updated to reflect #37
• #37 Studies now sort by time and date

0.3.7 (2014-02-25)¶

• #40 Restyled the filter section in the web interface and added a title to that section
• #38 Column titles tidied up in Excel exports
• #36 openrem_ptsizecsv output of log now depends on verbose flag
• #35 Numbers no longer stored as text in Excel exports

0.3.6 (2014-02-24)¶

• #34 Localised scripts that were on remote web servers in default Bootstrap code
• #33 Documentation now exists for adding data via csv file
• #24 Web interface has been upgraded to Bootstrap v3
• #5 Web interface and export function now have some documentation with screenshots

0.3.5-rc2 (2014-02-17)¶

• #32 Missing sys import bug prevented new patient size import from working

0.3.5 (2014-02-17)¶

• – Prettified this document!
• #31 Promoted patient size import from csv function to the scripts folder so it will install and can be called from the path
• #30 Improved patient size import from csv to allow for arbitary column titles and study instance UID in addition to accession number.
• #29 Corrected the docs URL in the readme

0.3.4-rc2 (2014-02-14)¶

• #28 XLSX export crashed if any of the filter fields were missing. Now fills on import with ‘None’
• #27 Use requested procedure description if requested procedure code description is missing

0.3.4 (2014-02-14)¶

• – General improvements and addition of logo to docs
• #23 Added Windows XP MySQL backup guide to docs
• #22 Added running Conquest as a Windows XP service to docs
• #15 Added version number and copyright information to xlsx exports
• #14 Added version number to the web interface
• #13 Improve the docs with respect to South database migrations

0.3.3-r2 (2014-02-04)¶

• #12 Added this version history
• #11 Documentation is no longer included in the tar.gz install file – see http://openrem.trfd.org instead

0.3.3 (2014-02-01)¶

Note

Installs of OpenREM earlier than 0.3.3 will break on upgrade if the scripts are called from other programs. For example openrem_rdsr is now called openrem_rdsr.py

• – Added warning of upgrade breaking existing installs to docs
• #10 Added .py suffix to the scripts to allow them to be executed on Windows (thanks to DJ Platten)
• #8 Removed superfluous ‘/’ in base html file, harmless on linux, prevented Windows loading stylesheets (thanks to DJ Platten)
• #7 Added windows and linux path examples for test SQLite database creation
• #6 Corrected renaming of example files installation instruction (thanks to DJ Platten)
• #4 Added some text to the documentation relating to importing files to OpenREM
• #3 Corrected copyright notice in documentation

0.3.2 (2014-01-29)¶

• Initial version uploaded to bitbucket.org

Release notes and upgrade instructions¶

Each release comes with specific upgrade instructions, so please follow the links below for the appropriate version.