Posts by gfcaetano

    How can I import an entire folder with subfolders with DICOM images into a Linux ConQuest server?
    Running "./dgate --addimagefile:file,patid" just adds 1 file into it, doesn't it?


    Thank you!

    Hi, syswiz.


    I'd like to know more about the way you do the ConQuest redundancy.
    Currently we have a backup team that does the backup like this:
    1) Today they get yesterday's studies to their workstations (with a local Conquest on their workstations);
    2) When the data folder stores almost 25GB they burn it on a BD-R disc labeling it with the studies date.


    My concern is about optical media reliability so I a want a different kind of backup. Sometimes studies aren't found on server and then we have to restore them by reading the BD-R discs, but sometimes we can't read some files from these discs anymore and the study is lost. I guess cloning constantly and automatically (to avoid human mistakes) our server would be the best backup method because if the main server crashes because of hardware issues we will be able to restore the system quickly using the backup clone server. But I'm not sure about the best way to do it.


    I don't trust the tech doing the manual push, a lot of times they just forget doing it.


    Have you used the ExportConverter at dicom.ini?
    We use this, but a sometimes (not always, so it works) we see ExportConverter failure messages at the logfile. Sometimes I know this happens because of network issues, sometimes I don't know why. But these studies will not be available at the backup server, or the ExportConverter tries again later?


    Regards!

    Hello.
    I've just installed the newest stable version of Conquest (1.4.16) on a Debian (stable - x86_64) server. It's working fine.
    My intention now is to use it as a Virtual Server for an old server, but without storing the old studies from the old server into the new server.
    I saw the OverlapVirtualGet at the manual, and it seems to do what I want, but I'm not sure how to set it up.


    The last line of my dicom.ini is like this:


    Code
    VirtualServerFor0 = CQCT


    CQCT is the AET of the old server.
    Where should I put the OverlapVirtualGet option? And which values for this option should I use?


    Thank you for helping!

    Hi, Marcel.
    We've enabled JPG support at our server and now we can successfully send the image to the server, but the image at our server has its colors changed.
    Using eFilm or K-PACS to send didn't work (the server shows "*** connection terminated"). We could send the images only with DCMSND (from DCM4CHE), but something is happening...


    We have a text page, with black text and in white background, saved as image and converted to DICOM. After sending it to our server, the background turns pink and the text turns dark green.
    After that we changed the image to grayscale (it was RGB) and sent it again to our server. It didn't have its "colors" changed, but when getting this image with eFilm or K-PACS, the displayed image is triple. We have tree views in one (side-by-side) of the same image.


    For now I don't have any more details, but if you ask me for some, I'll ask to the researchers and then tell you.


    Thank you for your attention.



    EDIT:


    The researchers found and corrected and error at their image conversion code, so the images are fine now! Thanks!

    ljxken, did you have your "Error saving to SQL" issue solved?
    I'm with the same issue here when a Siemens Emotion CT tries to send a specific study.

    Quote

    ***Error saving to SQL: 0879244I/1.3.12.2.1107.5.1.4.40202.4.0.714965037851816/1.3.12.2.1107.5.1.4.40202.4.0.714965475932048/1.3.12.2.1107.5.1.4.40202.4.0.7168104411846.dcm


    Thanks for any help!

    Hi, Marcel.


    Finally the server disks are full (only 31M free), so I've noticed that this command isn't doing what it should do:
    /opt/conquest/dgate -ff30000


    I've set up cron to run this command every night to keep 30GB of free space.
    I'm running it manually now and nothing is deleted (and nothing appears at "serverstatus.log". I still have only 31M free and can't receive any new studies.


    Now I'll try to delete old patient data so I'll be able to receive new ones.


    Thank you for any help.


    [ ]s!



    Edit: Now it worked, but only when entering '/opt/conquest' and then running './dgate -ff30000'
    I don't know why the disk got filled, because my cleanup script does this every night:


    cd /opt/conquest/
    ./dgate -ff20480 &



    ----------
    Editing...


    After analyzing what was going on, I've noticed that the server was receiving more than 20 GB of data per day. The disk was getting filled before the clean up script runs. Something went wrong with the server with the filled disk so it wasn't deleting before I restart it (when I wrote at this topic). After restarting the server and running './dgate -ff20480 &' again it freed 20GB.
    Now I've changed the -ff20480 parameter to -ff50960 and it's fine!

    Good morning.


    We're used to delete from server the patient with wrong ID and then repush the corrected data to server. But now we have another issue: the entered (wrong) PatientID has a space. When I try to remove this study I don't succeed.


    We've tried to remove using --deletestudy: and using the web (cgi-bin) page, but the study is still there.


    Is there any way to delete studies with spaces at PatientID field?


    Thanks!

    Hi!


    My friend is developing a computer-aided diagnosis system that automatically sends the post processing images to ConQuest server, but he is receiving this message when trying to send the images to the server:


    Quote

    WARNING:
    Abstract Syntax 1.2.840.10008.5.1.4.1.1.7 Secondary Capture Image Storage with
    Transfer Syntax 1.2.840.10008.1.2.4.50/JPEG Baseline (Process 1) not supported
    Cannot send c:\cad-pacs-client\result\1.2.40.0.13.1.10.165.41.57.13799530.20090911171058798.2


    It seems that ConQuest doesn't support this kind of syntax. Am I right?


    So my friend is asking the following question:


    Quote

    I'm not able to send images with Second Capture Image Storage transfer syntax to Conquest Server.
    Do you know if Conquest supports this transfer syntax?
    I appreciate you help. Thanks in advance!


    Thank you!


    PS: I'll ask him to join us at this forum.

    Sorry! My mistake!
    You're right, after your reply I saw that I was trying to run the maklinux_mysql from 1.4.14. I was at the wrong folder (1.4.14).


    Now I've just compiled 1.4.15alpha (using maklinux_mysql) and it's working!
    Everything I needed to do after compiling was to manually create the conquest database at MySQL.


    Is there any disadvantage of using ConQuest/MySQL under linux? Our Linux servers are using ConQuest/dBaseIII, but sometimes it crashes and when it restarts the db re-indexing takes a long time to finish. I guess I'll not have this issue using MySQL.
    Until 1.4.14, the file windowsmanual.pdf tells: "On Linux, one should use the built-in dbaseIII driver
    as long as the MySQL connection has not been developed."
    Was this the only reason to avoid using MySQL?


    Thanks again, Marcel!


    Regards.

    I'm running Debian (stable) at my server and after running maklinux_mysql (using 1.4.15alpha files) I see (using 'ls') that the dgate file is missing. It seems that something went wrong (too much warning messages, I couldn't scroll back to the beginning to find an error message) ...


    I have installed mysql-server (v5), apache (v2), g++ (v4.3), libpq-dev...


    Is anyone using ConQuest with MySQL with Linux?


    The content of maklinux_mysql is:


    Why there are some PostgreSQL references even if I will use MySQL? Is this correct? That's why I've installed libpq-dev.


    Thanks again!


    [ ]s!


    Quote from marcelvanherk

    Hi,


    It defitively is. There is a working? make script in 1.4.15alpha.


    Marcek

    Yes, Luishc. As Marcel said, the worklist server isn't used for images transfer.
    As far as I know, the worklist server is used at the scanner to get a "to-do" list of studies.


    Please, someone alert me if I'm saying something wrong.
    I didn't try Conquest as a Worklist Server yet, but this week I'll try to do it. I have only XML files, not HL7. I'm not sure if it will work, but if it doesn't I'll try to convert them to HL7. Maybe I'll use a batch file (loaded by cron every minute) to scan for new worklist items and add them to the Worklist Server (Conquest). Any better ideas?


    [ ]s!

    Hi!
    As stoneyaway didn't answer anymore I'll take this topic for myself. :)


    We're having this issue when we try to send one study (just this one, the others are fine) from our CR975 (Kodak) to our ConQuest 1.4.14 (Linux - dBaseIII):


    Quote

    Fri May 15 10:14:15 2009 ***Inconsistent PatientBir in DICOMStudies: PatientID = '0874047F' StudyInsta = '1.2.840.113564.101652289.2009042316020532890', Old='19990207', New='20080320'
    Fri May 15 10:14:15 2009 ***Refused to enter inconsistent link PatientID into DICOMStudies: PatientID = '0874047F' StudyInsta = '1.2.840.113564.101652289.2009042316020532890', Old='0898276B', Refused='0874047F'
    Fri May 15 10:14:15 2009 ***Error saving to SQL: 0874047F/1.2.840.113564.101652289.2009042316020532890/1.2.840.113564.101652289.2009042816081478183/1.2.840.113564.101652289.2009042816081478184.2003000225000.dcm


    What could be happening?


    Thank you!


    [ ]s!

    I just needed to read the results from "./dgate -help". I couldn't find that info before you answer me!


    Quote


    ./dgate -h


    ...


    [-ff#] Delete old patients until #MB free]


    Thanks again, Marcel.


    [ ]s!

    Well, I read completly the manual once, but I forgot that it has information about loading HL7 files. My mistake!


    Now, reading the manual again, I'm getting less confused! I still have some doubts, but I'll try some things on a test server to figure out how does it work!


    I'm sorry for asking something that was already answered, Marcel!
    And thanks (again)!


    [ ]s!

    Hi, everyone!


    As I told on some other topics, we use the Linux version of ConQuest (1.4.14) using the built-in dBaseII on a Debian (stable) system.
    I'm getting worried about the absence of the Windows interface of ConQuest because I don't know how to set the linux version to cleanup the disk automatically when it reaches some ammount of data.


    Is there any way to do it?
    Do any of you have any suggestions?


    Thanks a lot!


    [ ]s!

    Hello!
    Thanks again for all the support, Marcel!


    Here we're using ConQuest 1.4.14 for Linux (using dBaseIII) on a 4 TB server. It receives images from our CT, MR (Philips Achieva 3.0T), US, CR (Kodak - CR975)...
    Now we're trying to use a worklist server and I'm the man who has to make it work. Our CR will be the first scanner to read it.


    It's not clear to me the usage of ConQuest as a worklist server.
    You said "You can fill the worklist from the web interface or by dragging and dropping HL7 files", do I have to do it using Internet Explorer or any browser? It must be done at the server or it can be done using a client computer? Does it work when the server is running the Linux version?


    Thanks again!


    [ ]s!

    Hi, Marcel.
    Congratulations for this great server!


    I'm looking for a feature, I'm not sure that it is available already. I have a ConQuest server for MR studies, but some radiologists send CT studies to this MR server.
    Now we have hundreds of CT studies at the MR server. Is there any way to delete these CT studies from the MR server?


    Maybe using some command as dgate --deletestudy:"modalityuid".
    Is there a feature like this available?


    Thank you, Marcel!


    [ ]s!