Cross-sectional moment of inertia
Anita Barton
I'm fairly new at playing with ImageJ and BoneJ, and no-where near as
technically savvy as some of your other members!
For my PhD I am looking at cross-sectional CT images of equine MCIII's
before and after imposed exercise regimes, and was wondering if there
was a way to measure cross-sectional moment of inertia (moment of
area) in ImageJ? I suspect from some research it involves the 'slice'
function? From what I gather with the BoneJ Moment of Inertia
function, this requires a whole stack image?
Also, can I measure bone mineral density in a given R.O.I?
Many thanks!!
Anita
Michael Doube
For my PhD I am looking at cross-sectional CT images of equine MCIII's
It's a good bone to do a PhD on ;-)
was a way to measure cross-sectional moment of inertia (moment of
area) in ImageJ?
Yes, in BoneJ there is a plugin to do this:
http://bonej.org/slice
The doc is ambiguous about single-slice images, which should be handled properly (they are stacks with only one slice).
Also, the correct term for what you want is 'second moment of area'.
Also, can I measure bone mineral density in a given R.O.I?
Yes, but you will need proper calibration. We've discussed that here in previous posts, so have a look through old discussions. If you have pQCT images, watch this space.
Michael
Anita Barton
Thanks Michael, I'll have a play at work tomorrow. And yes, I should
have written SECOND moment of area ;-) I had printed out the 'slice'
page to play with (been busy teaching and data collecting this week).
I'll let you know if I'm not having success.
Can you clarify what you mean by the single stack images need to be
'handled properly'? Sorry, I'm very new at this!
Michael Doube
> Can you clarify what you mean by the single stack images need to be
> 'handled properly'?
BoneJ handles ordinary 2D images in Slice Geometry, no problem.
I put it in a confusing way - a 2D image can be considered a 3D image
with only one slice, or probably more accurately, 3D images are seen
by ImageJ as a series of 2D images ('slices'), so a single 2D image is
just a 3D image with only one slice.
> Sorry, I'm very new at this!
We all start somewhere!
Michael
Anita Barton
As I've said, I'm not very gifted with this stuff!
I've saved one of my CT dicoms as jpg, imported it into ImageJ. I
first tried to estimate SMOA just from that image untouched, but it
kept telling me there are no pixels available to calculate. I played
around with pixel value, but still no joy. Then I altered my image to
black bone with white background (using Image: adjust: threshold:
black and white), still no joy. I have also tried with and without
selecting a rectangular or circular ROI around my bone.
Help! :-)
Thanks!
Anita Barton
As I've said, I'm not very gifted with this stuff!
I've saved one of my CT dicoms as jpg, imported it into ImageJ. I
first tried to estimate SMOA just from that image untouched, but it
kept telling me there are no pixels available to calculate. I played
around with pixel value, but still no joy. Then I altered my image to
black bone with white background (using Image: adjust: threshold:
black and white), still no joy. I have also tried with and without
selecting a rectangular or circular ROI around my bone.
Help! :-)
Thanks
Michael Doube
> I've saved one of my CT dicoms as jpg, imported it into ImageJ.
Argh! Don't use JPEG! You lose data (it's called 'lossy' compression -
try zooming in so you can see the actual pixels) and the pixel values no
longer represent your real image data. ImageJ will open DICOMs, no
problem. It can open a whole series of DICOMs from a 3D CT scan, with
File->Import->Image Sequence, or you can just drag n drop a folder of
images onto ImageJ.
If you need to save processed images, use TIFF. The only use for JPEG is
to save bandwidth in an email or on the web. Don't use JPEG for image
processing. If you want to post an image to us on the list, use PNG
('non-lossy' compression) or the original file.
> first tried to estimate SMOA just from that image untouched, but it
> kept telling me there are no pixels available to calculate.
Sounds like a threshold or ROI problem.
> I played
> around with pixel value, but still no joy. Then I altered my image to
> black bone with white background (using Image: adjust: threshold:
> black and white), still no joy. I have also tried with and without
> selecting a rectangular or circular ROI around my bone.
Could you post an example to us (PNG or original) so we can try it out?
If you're not comfortable publishing your data, you can send it to me
off-list.
Michael
Anita Barton
Yes, I did notice my jpg's were not great quality for working in
imageJ :-/ However, ImageJ seems unhappy about opening my DICOMs from
my hard drive, it keeps telling me it cannot open the compressed
DICOM.
I have been opening them in Osirix, then saving them as a TIFF file,
and then convert to PNG on my desktop (Osirix wont save them as PNG,
and wont let me export DICOM onto my desktop)
I am still unable to get a SMOA with these files.
I'll send you a PNG file.
I can't see where to add it in here (I'm not usually THIS
untechnical), so I may have to post it off-list (although I have no
problems with it being on-list)
Cheers
Michael Doube
Apologies for my enthusiastic reply :-|
> it keeps telling me it cannot open the compressed
> DICOM.
http://www.loci.wisc.edu/bio-formats/imagej
There's more on the web if you Google "imagej compressed dicom"
> I have been opening them in Osirix, then saving them as a TIFF file,
> I'll send you a PNG file.
> I can't see where to add it in here (I'm not usually THIS
> untechnical), so I may have to post it off-list (although I have no
> problems with it being on-list)
email to bonej-users-a...@googlegroups.com
Michael
Anita Barton
Yes, the TIFF opens fine, just still telling me there are no pixels to
calculate, as before. I'm sure I need to play around with the
settings more. If I set the pixel values at -1500 to 1500 I do get a
table up, but I'm not convinced about the values.
I have emailed both the PNG and TIFF file for you. Look forward to
hearing your response! (make sure its enthusiastic (-;)