Welcome to the Podiatry Arena forums

You are currently viewing our podiatry forum as a guest which gives you limited access to view all podiatry discussions and access our other features. By joining our free global community of Podiatrists and other interested foot health care professionals you will have access to post podiatry topics (answer and ask questions), communicate privately with other members, upload content, view attachments, receive a weekly email update of new discussions, access other special features. Registered users do not get displayed the advertisements in posted messages. Registration is fast, simple and absolutely free so please, join our global Podiatry community today!

  1. Have you considered the Clinical Biomechanics Boot Camp Online, for taking it to the next level? See here for more.
    Dismiss Notice
Dismiss Notice
Have you considered the Clinical Biomechanics Boot Camp Online, for taking it to the next level? See here for more.
Dismiss Notice
Have you liked us on Facebook to get our updates? Please do. Click here for our Facebook page.
Dismiss Notice
Do you get the weekly newsletter that Podiatry Arena sends out to update everybody? If not, click here to organise this.

Osirix

Discussion in 'General Issues and Discussion Forum' started by drsarbes, Dec 16, 2008.

  1. drsarbes

    drsarbes Well-Known Member


    Members do not see these Ads. Sign Up.
    Anyone using the software OSIRIX to read MRI and CT scan?

    If so, is there anyway to get 3D rendering (non surface) of the foot/ankle form an MRI or CT?

    Steve
     
  2. Lee

    Lee Active Member

    Hello Steve,
    Do you mean Osiris? I used a freeware program called Osiris for charting digital X-Rays in pre and post HAV surgery for a masters thesis. At the time, there were no 3d rendering functions available on it. (Oh, I just looked it up - seems the same people are now producing Osirix - but I can't seem to access it off their website).
    What do you want to create 3D renderings for?
    I have previously used a few different programs for 3D reconstruction of MR and CT scans for research purposes. There's a freeware one called 'Slicer' (http://www.slicer.org/), which I found to be ridiculously long winded and quite unstable. Then there's the rather expensive Amira (be really easy to use) - http://www.amiravis.com/. I also used Mimics (http://www.materialise.com/materialise/view/en/92458-Mimics.html), which was a compromise between the others.
    Cheers,
    Lee
     
  3. drsarbes

    drsarbes Well-Known Member

    Hi Lee:
    Being a MAC GUY, I gravitated to Osirix to view DICOM files when things went digital. I was getting MRI, etc....on CD and couldn't view them. Osirix cured all that and then some. FRom what I understand this FREE software is superior to software costing thousands.

    I have seen vascular imaging rendered in 3D...... I can do this for CT scans, of the skeletal structures, of course.

    I was wondering if this is possible for MRI at some level other than surface rendering, which I find useless. It would be nice to be able to STRIP away layers of structures from skin to bone in a 3D fashion using DICOM files from an MRI.

    Steve
     
  4. Lee

    Lee Active Member

    Hello Steve,

    Yes, I'm sure it will be, but most of the software I've used is really reliant on at least some partial manual delineation of tissue planes slice-by-slice. The more slices you have, the better the resolution of your final 3D image, but the longer it takes you! I found most clinical MR and CT scans to have insufficient number of slices to give you quality 3D renderings of foot bones (even the calcaneus!). If you've access to MR and a friendly technician, they will be best at giving you advice on how to improve resolution (eg. using nice small wrist coils and higher tesla rated scanners for MR), but at some stage you will be making a compromise between scan area, scan times, amount of slices and resolution. Most imaging software will rely on contrast to make delineation of tissue planes possible, and since soft tissues are varying shades of grey on MR, I think this will be time consuming and take a bit of guesswork on the operator's part to an extent.

    The old Osiris program had some volume rendering capabilities, and had demo's on vascular stuff, but was pretty useless for bone, etc.. Maybe they've improved it with Osirix - and I hope they improved their instruction manual and user interface!

    Unless there's some other software out the that can automate this process? If there is please let me know!

    Cheers,
    Lee
     
  5. drsarbes

    drsarbes Well-Known Member

    "Maybe they've improved it with Osirix - and I hope they improved their instruction manual and user interface! "

    WHAT MANUAL????
    Blog, message boards and a "how to" video is all I can find.
    Thanks for the input. I understand.
    Steve
     
  6. drsarbes

    drsarbes Well-Known Member

    Hi Lee:
    I've done some more searching and have discovered that the 3D volume rendering doesn't work well (period) for MRIs -
    It works great with CT............ but because I seem to order more MRIs I was hoping there was a way to perhaps change density values and do the 3D thing for the dicom MR data.
    Guess not.
    If anyone out there knows a way to do this, I would greatly appreciate some info.

    Thanks

    Steve
     
  7. Woodburn

    Woodburn Active Member

    Dear Steve,
    I've used software called 3DVIEWNIX when I was a visiting scholar at the Medical Imaging Processing Group at U.Penn (http://www.mipg.upenn.edu/). Prof. Jay Udupa's team has developed particular expertise in developing highly accurate and precise segmentation algorithms. I have used both MRI and CT data acquisitions for foot and wrist joints in rheumatoid arthritis (see references below). The software has powerful 'live-wire' segmentation which is obviously automated after manually depositing the first seed point. In MRI I have successfully segmented contrast-enhanced regions in bone, joint and tendon indicative of synovitis, tendinopathy and bone oedema as well as full bone segmentation and 3D volume reconstructions. I still use the software mounted on a linux platform. The bone/joint reconstructions from CT are obviously better than MRI but it may be possible to register the images from dual acquisition and then overlay soft-tissue.


    Woodburn J, Udupa JK, Hirsch BE, Wakefield RJ, Helliwell PS, Reay N, O’Connor P, Budgen A, Emery P. The geometrical architecture of the subtalar and midtarsal joints in rheumatoid arthritis based on MR imaging. Arthritis Rheum 2002;46:3168-3177.

    Goldbach-Mansky RP, Woodburn J, Yao L, Lipsky P. Magnetic Resonance Imaging in the evaluation of bone damage in patients with Rheumatoid Arthritis. Arthritis Rheum 2003;48:585-9.

    Udupa JK, LeBlanc VR, Zhuge Y, Imielinska C, Schmidt H, Currie LM, Hirsch BE, Woodburn J. A framework for evaluating image segmentation algorithms. Comput Med Imaging Graph 2006;30:75-87.

    Kind regards,

    Prof. Jim Woodburn
    Professor of Rehabilitation
    Foot & Ankle Research Group,
    HealthQWest,
    Glasgow Caledonian University,
    Glasgow,
    Scotland.

    jim.woodburn@gcal.ac.uk
     
  8. drsarbes

    drsarbes Well-Known Member

    Hi Professor Jim:
    Thank you for the information, I'll check into it.
    Are you aware of a 3D volume rendering from non contrast MRIs.
    I guess what I'm asking (and forgive my ignorance) is if there is a way to, for instance, specify densities to tendon structures and render a 3D image as you might from a contrast image.

    Thanks again

    Steve
     
  9. Woodburn

    Woodburn Active Member

    Dear Steve,
    Likewise I'm not an expert! Certainly in 3DVIEWNIX you have different segmentation options from live-wire to manual as well as thresholding. Both live-wire and thresholding with fuzzy segmentation have worked well to seperate out inflamed from normal structures in RA. Its one of things you have to play with. If you can remove your header information from your DICOM files I could always have a look for you (you can download shareware software that would remove header information).
    Best wishes,
    Jim
     
  10. drsarbes

    drsarbes Well-Known Member

    Hi Jim:
    I exported a 3D vol rendering movie of an MRI, it's only 2.6MB in a quicktime mov. format.
    Apparently I can't upload it here. If you could send your e-mail address to me at drsarbes@aol.com I could send it to you. Perhaps you could instruct me on how to strip some layers away.
    Thanks
    Steve
     
  11. Lee

    Lee Active Member

    Hello Steve,

    It's unlikely that you'd be able to do a great deal of image manipulation with a movie. Really, you need to keep the DICOM files in their original format (and remove the headers like Jim suggested) before putting them into one of the 3D reconstruction programs. I think this is possible with a program called irfanview and one of its add-ons - http://www.irfanview.com/

    If not, just google freeware DICOM software until you find one that's useable.
    Jim's work was novel (and sparked my original interest in 3D reconstruction of foot and ankle pathology) and provided quantification of synovitis and bony architecture - these structures are relatively (and I use that with reference to how difficult I found 3D reconstruction of non-pathological soft tissue) easy to reconstruct due to their contrast to the surrounding tissues. As you'll know, MR provides a grey scale image based on relative water (sort of, but not really) content of tissues. Anything that's inflamed will have a better contrast to surrounding tissues - a great example of this is osteomyelitis tracking up a metatarsal (or synovitis as in some of Jim's work). Bone is generally poorly demonstrated on MR, but the contrast between it and surrounding soft tissue is pretty clear - so it's quite easy to reconstruct.

    Soft tissue structures with very similar tissue components (eg. collagen fibres in ligaments and tendons) have relatively poor contrast in non-pathological subjects. This is where thresholding functions in any of these programs may let you down. It might be easier to decipher muscular compartments than tendons & ligaments and their origin and insertion. It's a difficult one really, might be worth speaking to your local Radiologist (especially if they're academically inclined) or contact one of Jim's collaborators?

    Merry Christmas,
    Lee
     
  12. Woodburn

    Woodburn Active Member

    I agree with Lee that it would be extremely difficult to segment out the small ligaments of the foot with anything other than a manual technique which I suspect would have poor accuracy and precision. Moreover, your scans slices are 3mm thick so your highly vulnerable to partial volume effects, something that automated segmentation algorithms just don't like. Unless you focusing on one or two structures related to specific condition or injury this would be a very time consuming and laborius process. Our work with 3DVIEWNIX has yet to be translated into routine clinical use.
    Best wishes for a very merry Christmas,
    Jim
     
  13. drsarbes

    drsarbes Well-Known Member

    Thanks Jim for looking at the movie I sent you.

    And thank you Lee as well.

    I wasn't trying to manipulate the movie, really, it was just the format that I saved the rendering in.

    Thanks to you both I'm beginning to understand the inherent difficulty in rendering a 3D Image for a select anatomic part from MRI.

    Steve
     
  14. Lee

    Lee Active Member

Share This Page