DICOM server 1.4.19 released

Tonight, at last, version 1.4.19 of the conquest DICOM server has been released.

You can find it here: https://ingenium.home.xs4all.nl/dicom.html

I would like to thank all users for feedback and error reports. Please keep on doing this to allow improving the quality of the software.

The latest updates include: A complete recoding of the browser in the GUI for speed and lots of new options such as flexible DICOM modification, server speedup, JpegLS and better Jpeg2000 compression (thanks Bruce Barton), a completely scripted web interface with a new interactive viewer, and options for a new flexible anonymisation server.

Marcel

Here bug reports for 1.4.19 will be collected.

These are leftover bugs:
1) add LittleEndianExplicit to GUI connecting to dgate. Futured
2) Improve safety of sqlite by regular flushing. Futured
3) Check these changes: 20140608 mvh Added BYTE PCid to CStoreRSP::Write; This change is under review: pcid does not seem to get the right value
4) brainix sample data fails to load from zip; maybe because of French folder names
5) sql server and access database ODBC install only work for 32 bits of dgate (e.g. disable 64 bits or create ODBC entry by hand)
6) Autoinstall newweb - this is in the web based installer in 1.4.17a
7) install second server when service of first runs gives trouble - do an echo on the install button and fail if found - futured
8 ) last line of dicom.ini does not read if no CR (IndexDBF = 50) - futured
9) Publish about Docker release of Lance
10) gui freeze on window 10 - not reproduced so far
11) GUI fails to run as user after install as service on server2012; does run as admin - futured
12) Implement wadoviewerhelp - futured
13) Implement web interface for batch anonymisation - futured
14) overlapvirtualget=0 breaks the virtual server operation (hardw) - futured
15) Zoom disabled for wadoviewers for version 1.4.19 < 1.4.19b (4 files) - in 1.4.19a
16) Thumbnail in MR confuses kpacs viewer (Dennis on mail) - futured
17) weasis starter has 127.0.0.1 hardcoded, must be %serveraddress%
18) wish: OperatorConsole.printf ("*** multiplex: connection terminated [%s]\n", CallingApTitle);
19) c-move response cannot convey >65536 images, should max numbers at 65535 - futured

20) I updated about 12h after first release to properly use charls1.x; please use release file dated 9 February. In 1.4.19a
21) After Linux install, cgi-bin subfolders such as newweb have insufficient access rights. Fix with sudo chmod 777 /usr/lib/cgi-bin/newweb and same for viewers folder. In 1.4.19a
22) Fixes needed to MySQL Linux documentation. Use sudu apt-get install libmysql++-dev instead of libmysql-client-dev; mysql --user=root --password=xxx, see below. In 1.4.19a
23) Fixes needed to MySQL Linux build (maklinux). Add $SRC/ before jpeg-6c in in line 29 (2 times, see below). In 1.4.19a
24) newweb interface displays lua error when UIDMods does not yet exist (in line return #a, suggest to change to #(a or {})). Does not affect operation. In 1.4.19a
25) for anypage to block default web interface, also set defaultpage
26) default install causes some issues. Use manual install. - futured
27) color print server does not work. - futured, awaiting more information.
28 ) 1.4.19 does not seem to run on Windows XP; special XP compile with MS100 seems unstable. Need to debug it.
29) servercommand may need larger bufferspace for return data for future service web page. In 1.4.19a
30) Move result and merged_row out of HAVE_LIBJPEG to ToJPGStrip to allow compile without jpeg-6c. In 1.4.19a
31) remove typedef INT32 in jpeg_encoder.cpp. In 1.4.19a
32) There is no Lua call for servercommand to console, needed for generating large web items (e.g. zip); suggest extra par '', 'cgibinary', 'cgihtml', filename. In 1.4.19a
33) lua servercommand cannot return binary data as HTML processing Is enabled by default. In 1.4.19a
34) dicommove documentation is incorrect, if 4th parameter is passed 1 means patientroot move, else studyroot; In 1.4.19a
35) merge series button is broken in GUI; request to add merge studies button. In 1.4.19a
36) Modifystudy seems to have a problem (not reproduced)
37) Lossless jpeg compressed color US image decompressed green (lost 2014 fix). In 1.4.19a

Please report all problems as this will improve future releases. Thanks!

Marcel

Here feature requests for 1.4.19 will be collected

I just tested the 1.4.19 release with MySQL on Linux and found some flaws. Here is what to do.

install g++ and apache etc as per linuxmanual

Then:

sudo apt-get install mysql-server
(set and remember root password, here for example: as_set_above)
sudo apt-get install mysql-client
sudo apt-get install libmysql++-dev
gedit maklinux
on line 29, add $SRC/ before jpeg-6c (2 times)
save it

run ./maklinux option 1
mysql --user=root --password=as_set_above
create database conquest;
exit;
edit dicom.ini to set Password = as_set_above

./dgate -v -r
./dgate

Now everything runs!

Marcel

Hi,

Thamks for the report, I will have a look soon. Did you try manual install? I did test that.
What database did you select?

Marcel

Hi,

apparantly default install pushes the 'create data source' button even when it is disabled (such as for sqlite). That may cause the error. After the error, dgate64 or conquestdicomserver may still run in the background, blocking access. Try again with normal install.

Marcel

Hi,

Thanks for the report. I have not tried on XP for a while, develop on 7, test on 8 and 10 and Linux. My best bet would be to combine conquestdicomserver.exe from the last 1.4.17 with the dgate(64).exe from 1.4.19. This should run fine and offer all features of the new server (not the GUI though).

regards,

Marcel

Could be,

I was forced to switch compilers for both exe's and they may depend on libraries that XP doesn.t have. I will have a quick look, but it may be hard to support XP.

regards,

Marcel

Will have a look. The Vitrea fix should be in 1.4.19.

Marcel

Hi,

The issue is in dgate(64).exe due to the VS2015 compiler. I can compile for XP using an older compiler (MS100) and a few changes, but is does not seem to run reliably. I attached the 2 exe's for you to test.

Marcel

Hi,

you can write a script to delete the data on such occurence, but you would need to delete the whole series. But conquest has no better option than what it does now, see post "why not chnage patientID of anm image". By the way, changing name only would not get a refusal. It is due to changing the patientID.

Have a look at RejectedImageConverter0 in the manual and command 'retry', e g.,

RejectedImageConverter0 = newuids; retry

regards,

Marcel

Yep. But it may delete too much, be careful and try it out on an empty testserver first. You add these lines to dicom.ini. It may not work at all (untested).

RejectedImageConverter0 = delete study; retry

you can also experiment with delete series or delete image. I have a feeling though the delete command may be broken.

Or try (lua equivalent, more likely to work), add at bottom of dicom.ini

[lua]
RejectedImageConverter0 = x=DicomObject(); x.SeriesInstanceUID = Data.SeriesInstanceUID; dicomdelete(x); script('retry')

be careful!

Marcel