Hi Yi,<br><br>Currently we don't have a program for what you need. We use our own codes in IDL or SPM (MatLab). Ken and Andreia are doing it routinely. Please ask them if you have specific questions. There are a couple of points I want to make;<br>
<br>1) If you normalize your data to an atlas, all your data will have similar shape. Then, you can perform pixel-by-pixel statistics. This is voxel-based analysis. I believe SPM has an array of tools for this. <br><br>2) If you transform an atlas (especially our white matter parcellation map (WMPM)) to the shape of your data, you will get numbers (such as volume, FA, MD, etc) from 100-176 regions of each brain (automated segmentation). This process converts image data to a string of numbers. Then the data can be analyzed by Excel for statistical analyses. <br>
<br>The approach #1 is ideal for strong abnorality in a small region. The approach can theoretically detect abnormality as small as one pixel but each pixel has low statistical power.<br><br>The approach #2 is ideal for diffuse abnormality. It groups all pixels that belong to one of the 100-176 regions, which increases the statistical power but if the abnormal region is only a small portion of one segment, the pixel grouping reduces the sensitivity. <br>
<br>So, the two approaches are complementary. We have been working on a statistical tool for approach #1 but it'll take some time.<br><br>Susumu<br><br><div class="gmail_quote">On Thu, Aug 6, 2009 at 2:32 PM, Yi Jiang <span dir="ltr"><<a href="mailto:yj3@duke.edu">yj3@duke.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div bgcolor="#ffffff">
<div><font size="2" face="Arial">Hi Dr Mori,</font></div>
<div><font size="2" face="Arial"></font> </div>
<div><font size="2" face="Arial">Thank you very much. Andreia and Xin have helped me
to make it work.</font></div>
<div><font size="2" face="Arial"></font> </div>
<div><font size="2" face="Arial">I have another question. I am wondering if you
have suggestion or software about averaging eigenvectors or tensors. In
other words, if we have, say, DTI dataset of 10 brains, all registered together,
how can we calculate the mean (and variation) of the 10 primary eigenvector
field or calculate the mean (and variation) of the 10 tensor
field?</font></div>
<div> </div>
<div><font size="2" face="Arial">Any suggestion is highly appreciated!</font></div>
<div> </div><font color="#888888">
<div><font size="2" face="Arial">Yi</font></div>
<div> </div>
</font><blockquote style="border-left: 2px solid rgb(0, 0, 0); padding-left: 5px; padding-right: 0px; margin-left: 5px; margin-right: 0px;"><div class="im">
<div style="font-family: arial; font-style: normal; font-variant: normal; font-weight: normal; font-size: 10pt; line-height: normal; font-size-adjust: none; font-stretch: normal;">----- Original Message ----- </div>
<div style="background: rgb(228, 228, 228) none repeat scroll 0% 0%; font-family: arial; font-style: normal; font-variant: normal; font-weight: normal; font-size: 10pt; line-height: normal; font-size-adjust: none; font-stretch: normal; -moz-background-clip: -moz-initial; -moz-background-origin: -moz-initial; -moz-background-inline-policy: -moz-initial;">
<b>From:</b>
<a title="susumu@mri.jhu.edu" href="mailto:susumu@mri.jhu.edu" target="_blank">susumu mori</a>
</div>
</div><div><div></div><div class="h5"><div style="font-family: arial; font-style: normal; font-variant: normal; font-weight: normal; font-size: 10pt; line-height: normal; font-size-adjust: none; font-stretch: normal;">
<b>To:</b> <a title="yj3@duke.edu" href="mailto:yj3@duke.edu" target="_blank">Yi Jiang</a> ; <a title="mristudio-users@mristudio.org" href="mailto:mristudio-users@mristudio.org" target="_blank">DTI Studio, ROI Editor, Landmarker
Questions/Support</a> ; <a title="xli16@jhmi.edu" href="mailto:xli16@jhmi.edu" target="_blank">Xin Li</a> ; <a title="afaria1@jhmi.edu" href="mailto:afaria1@jhmi.edu" target="_blank">ANDREIA FARIA</a> </div>
<div style="font-family: arial; font-style: normal; font-variant: normal; font-weight: normal; font-size: 10pt; line-height: normal; font-size-adjust: none; font-stretch: normal;"><b>Sent:</b> Tuesday, August 04, 2009 9:59
PM</div>
<div style="font-family: arial; font-style: normal; font-variant: normal; font-weight: normal; font-size: 10pt; line-height: normal; font-size-adjust: none; font-stretch: normal;"><b>Subject:</b> Re: [Mristudio-users] how to
save the transformation matrix from alandmarker based transformation</div>
<div><br></div>
<div>Hi Yi,</div>
<div> </div>
<div>I believe you can save the matrix. Xin should know how to do it.</div>
<div>As for the matrix combination, I remember that we discussed about exactly
the same topic in my lab. I thought that answer was Yes, but don't exactly
remember how to do it. Andreia and Xin, do you know how to do
it?<br><br></div>
<div class="gmail_quote">On Tue, Aug 4, 2009 at 4:31 PM, Yi Jiang <span dir="ltr"><<a href="mailto:yj3@duke.edu" target="_blank">yj3@duke.edu</a>></span>
wrote:<br>
<blockquote style="border-left: 1px solid rgb(204, 204, 204); margin: 0px 0px 0px 0.8ex; padding-left: 1ex;" class="gmail_quote">
<div bgcolor="#ffffff">
<div><font face="Arial">Dear All,</font></div>
<div><font face="Arial"></font> </div>
<div><font face="Arial">I have tried to look through the online manual, but
could not find the answer.</font></div>
<div><font face="Arial"></font> </div>
<div><font face="Arial">I tried to use the landmark based transformation
(translation) from the main menu of the Landmarker window. The result
is pretty good. However, I do not know how to save the transformation
matrix. I tried to click the "save transformation matrix" button in the
subject window. It seems two files was saved (.hdr? and .dat), but the file
sizes were very small.</font></div>
<div><font face="Arial"></font> </div>
<div><font face="Arial">I also want to combine the transformation matrix from
the landmark based translation with a AIR linear rigid, a AIR linear
affine transformation, and a LDDMM matrix. Can I do that? I do not know
if the transformation matrix from landmark based translation is a AIR
format, or LDDMM raw format, or a LDDMM vtk format, which was asked in the
combine matrix window.</font></div>
<div><font face="Arial"></font> </div>
<div><font face="Arial">The last question may be stupid: if there are
defined landmarks in the memory, when I click on AIR linear or multichannel
LDDMM (not from the main menu, but the buttons in the subject window), those
landmarks were not used in transformation, right?</font></div>
<div><font face="Arial"></font> </div>
<div><font face="Arial">Thank you very much!</font></div>
<div><font face="Arial"></font> </div>
<div><font face="Arial">Best,</font></div>
<div><font face="Arial">Yi</font></div>
<div><font face="Arial"></font> </div>
<div><font face="Arial"></font> </div>
<div><font face="Arial"></font> </div></div><br>_______________________________________________<br>Mristudio-users
mailing list<br><a href="mailto:Mristudio-users@mristudio.org" target="_blank">Mristudio-users@mristudio.org</a><br><a href="http://lists.mristudio.org/mailman/listinfo/" target="_blank">http://lists.mristudio.org/mailman/listinfo/</a><br>
Unsubscribe,
send a blank email to: <a href="mailto:Mristudio-users-unsubscribe@mristudio.org" target="_blank">Mristudio-users-unsubscribe@mristudio.org</a><br><br></blockquote></div><br></div></div></blockquote></div>
<br>_______________________________________________<br>
Mristudio-users mailing list<br>
<a href="mailto:Mristudio-users@mristudio.org">Mristudio-users@mristudio.org</a><br>
<a href="http://lists.mristudio.org/mailman/listinfo/" target="_blank">http://lists.mristudio.org/mailman/listinfo/</a><br>
Unsubscribe, send a blank email to: <a href="mailto:Mristudio-users-unsubscribe@mristudio.org">Mristudio-users-unsubscribe@mristudio.org</a><br>
<br></blockquote></div><br>