Use Conquest to convert Bit depth

    Hi Mukoya,


    with lua scripting you can read and scale pixels row by row, but it will be slow (a few seconds per 512x512 image). See sample file distxfm.lua for inspiration. It may also be that changing BitsStored for your data is enough. What is the problem you are trying to solve?


    Marcel

    Thank you very much for your reply.


    Well, i am new to Conquest and amazed at what it has to offer!


    Anyway, I would like to change image bit depth so I can have the studies compressed with JPEG lossy without need to use j2k. Of cause for this, changing BitsStored does not do the trick (I tried it before)


    I still do not understand lua at all. i will have to take some time trying to understand it then i can see what to do about bit depth.


    I have a Toshiba CT equipment: Aquilion One 640 slice. produces 16 bit images and vendor says it cannot be changed to 12bit by the equipment. Been using clearcanvas server which wont do j2k compression.


    One way to solve this is to have a rule that deletes all series containing series description "Vol". This is because storage is a major issue especially if lossy compression not possible.


    A CT abdomen with two original series will send two Volume series each with 15000-2500 axial images. The equipment will also automatically send reconstructed images e.g 120 axials per volume series plus about 50 Sagittal and 50 Coronals for each volume series. Volume series can have more than 4000 images e.g peripheral angio.


    The idea is to delete the volume series after reporting, say after 3 months by passing the study through Conquest, yet retaining the reconstructed series. All volume series have "Vol." in their series description.


    best regards.

    Hi,


    there is an issue in the current conquest version 1.4.17b that affects lossy compression, which will not be activated at all for 16 nor 12 bits images. In the upcoming fix changing BitsStored to 12 should do the trick, but I am not sure what you would lose.


    Deleting specific series on date should be fairly easy with a lua script. Samples are on the forum and in the 1.4.17b distribution.


    Nice data sets by the way!


    regards,


    Marcel

    Thank you very much for prompt reply.


    I have attached a sample screenshot identifying the series I would like to delete from this sample study. I have also noticed that I might need to use another filter e.g Contains "Vol" and has more that 100 instances etc. This is because, as you can see in the screen shot, some 3D images also have a series description similar to the volume.


    I will do some research about lua to get more educated then try it out. I think it would be easier to seek help then!


    Well I am having trouble attaching the screenshot. I might do it later. Thanks all the same.


    mukoya.

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!

Register Yourself Login