Measuring Volumes from MRI images

[Seth Shipman]

Hi, I'm trying to use ImageJ to get hippocampal volumes from structural MRI images.  I can get them by tracing the HPC on each slice, using the measure function to get area, exporting the data to excel and calculating volume from all of the areas, but I'm wondering if anyone is doing this with a more sophisticated technique.  I would appreciate any suggestions for specific plugins or tools that I might be missing.  Thanks,

Seth Shipman




-----------------------------------------------
Confidentiality Notice

This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential or proprietary information which is legally privileged.  Any unauthorized review, use, disclosure, or distribution is prohibited.  If you are not the intended recipient, please promptly contact the sender by reply e-mail and destroy all copies of the original message.

[Martin Wessendorf]

Seth Shipman wrote:

> Hi, I'm trying to use ImageJ to get hippocampal volumes from structural MRI images.  I can get them by tracing the HPC on each slice, using the measure function to get area, exporting the data to excel and calculating volume from all of the areas, but I'm wondering if anyone is doing this with a more sophisticated technique.  I would appreciate any suggestions for specific plugins or tools that I might be missing.  Thanks,

There's the grid plugin that lets you do point-counting instead of
measuring the area but I'm not sure it provides much of an advantage one
way or the other.  In any case, before you begin, find a copy of
"Unbiased Stereology" by CV Howard and MG Reed, or "Unbiased Stereology"
by PR Mouton.  They'll walk you through the sampling issues, which turn
out to be more important than the precise method by which you measure area.

Good luck!

Martin Wessendorf
--
Martin Wessendorf, Ph.D.                   office: [phone]
Assoc Prof, Dept Neuroscience                 lab: [phone]
University of Minnesota             Preferred FAX: [phone]
6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
Minneapolis, MN  55455                E-mail: [hidden email]

[Albert Cardona]

Seth,

I created a plugin in the past for modelling that will give you the
estimated volume as well ( http://www.pensament.net/java/ ). I must
admit I have seldom used the volume measuring, so I don't know how
precise it is, some tests are advisable. In any case you can get the 3D
model out of it, from which perhaps CAD programs can measure the volume
for you.

Albert

[Robert Dougherty]

There is also the misnamed Measure Stack
http://www.optinav.com/ImageJplugins/MeasureStack.htm

Bob

Robert P. Dougherty, Ph.D.
President, OptiNav, Inc.
Phone [phone]
Fax [phone]
www.optinav.com
 

> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> Albert Cardona
> Sent: Wednesday, November 30, 2005 11:31 PM
> To: [hidden email]
> Subject: Re: Measuring Volumes from MRI images
>
> Seth,
>
> I created a plugin in the past for modelling that will give you the
> estimated volume as well ( http://www.pensament.net/java/ ). I must
> admit I have seldom used the volume measuring, so I don't know how
> precise it is, some tests are advisable. In any case you can get the 3D
> model out of it, from which perhaps CAD programs can measure the volume
> for you.
>
> Albert

[ToeKnee Lacey]

Hi,

You might want to look at the <a href="http://ip.beckman.uiuc.edu/Software/SplineSnake/">

spline snake</a> plugin.

I've also developed some tools that we use to
manually markup cartilage knees from animal MRI data.  It's grown into quite
a sophisticated plugin allowing the user to identify the boundary over two
stacks simultaneously (pre and post treatment) whilst controlling the interations
between the two stacks to ensure that certain shape consistencies are enforced.


To get this going I had to overcome a couple of limitations with ImageJ.
 One was that it doesn't display images in their 'true' aspect ratio but instead
assumes pixels are square.  As we know this is often untrue in MRI.  The second
was the lack of subpixel mouse position information which is critical for
marking up small structures like these.  These I fixed by extending thes canvas
in imagej.  However this does require a Java 2 system.

I've also integrated
some code to communicate with an <a href="http://www.r-project.org/">R server</a>
so that imagej can squirt results directly to this excellent open source statistics
package either running locally or on a remote server (which can then do more
complex stats, generate webpage output etc).

I've developed some Matlab/Java
code to deserialise Java objects containing imagej structures for further
analysis in Matlab.  Matlab  runs a Java virtual machine so this is a relatively
simple process (once you fix the classloader!) and I find it another great
way to do further analysis on markups without all the bother of writing complex
file readers/writers or  compromising the design of your classes.  Again I
think this has generic utility.

Apologies for going off track a bit but
I'm interested to know if this stuff would be of wider utility.  Although
the specifics of the plugin are  geared towards our problem I think the framework
is a sound one and there are some bit here I haven't seen elsewhere.

-tony


--- ImageJ Interest Group <[hidden email] wrote:
Hi, I'm trying
to use ImageJ to get hippocampal volumes from structural MRI images.  I can
get them by tracing the HPC on each slice, using the measure function to get
area, exporting the data to excel and calculating volume from all of the areas,
but I'm wondering if anyone is doing this with a more sophisticated technique.
 I would appreciate any suggestions for specific plugins or tools that I might
be missing.  Thanks,
>
> Seth Shipman
>
>
>
>
> -----------------------------------------------

> Confidentiality Notice
>
> This e-mail message, including any attachments,
is for the sole use of the intended recipient(s) and may contain confidential
or proprietary information which is legally privileged.  Any unauthorized
review, use, disclosure, or distribution is prohibited.  If you are not the
intended recipient, please promptly contact the sender by reply e-mail and
destroy all copies of the original message.
>

[Sankar, Andiappa]

Hi,

  If I am understood correctly, you are interested to import the dose matrix / fluence map to Image J. My experience is with  Varian's Eclipse TPS Files.

   (1) I did import the dose matrix file [dicom]  through the following option,

                  File - Import - Image sequnence , with "8-bit Gray scale option" selected.

   (2) If u want to import the fluence map files , you have to delete the header informations using notepad / wordpad
       and import the file as a TEXT IMAGE.

   If u need example files, please mail to  "[hidden email]"

    Regards,
Sankar.

*****************************************************************
The information contained in this message may be confidential or
legally privileged and is intended for the addressee only, If you
have received this message in error or there are any problems,
please notify the originator immediately. The unauthorised use,
disclosure, copying or alteration of this message is
strictly forbidden.
*****************************************************************

[Gerald Veldhuijsen]

Hi,

thank you.
No, it turned out that my problem was that I was trying to import
compressed images. When I tried an uncompressed sequence it worked ok.
The problem is that currently ImageJ doesn't support lossless JPEG
compression.

Regards,

Gerald










"Sankar, Andiappa" <[hidden email]>
Sent by:
ImageJ Interest Group <[hidden email]>
12/05/2005 11:32 AM
Please respond to
ImageJ Interest Group <[hidden email]>


To
[hidden email]
cc

Subject
Import dose structures in Image J
Classification







Hi,

  If I am understood correctly, you are interested to import the dose
matrix / fluence map to Image J. My experience is with  Varian's Eclipse
TPS Files.

   (1) I did import the dose matrix file [dicom]  through the following
option,

                  File - Import - Image sequnence , with "8-bit Gray scale
option" selected.

   (2) If u want to import the fluence map files , you have to delete the
header informations using notepad / wordpad
       and import the file as a TEXT IMAGE.

   If u need example files, please mail to  "[hidden email]"

    Regards,
Sankar.

*****************************************************************
The information contained in this message may be confidential or
legally privileged and is intended for the addressee only, If you
have received this message in error or there are any problems,
please notify the originator immediately. The unauthorised use,
disclosure, copying or alteration of this message is
strictly forbidden.
*****************************************************************

[Marek Maryanski]

thank you so much for the info and offer
  yes
  i would love to test this on an example
  if you can
  please send me some examples
  for me to practice on
  thanks again
   
  marek

"Sankar, Andiappa" <[hidden email]> wrote:
  Hi,

If I am understood correctly, you are interested to import the dose matrix / fluence map to Image J. My experience is with Varian's Eclipse TPS Files.

(1) I did import the dose matrix file [dicom] through the following option,

File - Import - Image sequnence , with "8-bit Gray scale option" selected.

(2) If u want to import the fluence map files , you have to delete the header informations using notepad / wordpad
and import the file as a TEXT IMAGE.

If u need example files, please mail to "[hidden email]"

Regards,
Sankar.

*****************************************************************
The information contained in this message may be confidential or
legally privileged and is intended for the addressee only, If you
have received this message in error or there are any problems,
please notify the originator immediately. The unauthorised use,
disclosure, copying or alteration of this message is
strictly forbidden.
*****************************************************************