Hi Kirles,<div><br></div><div>Luis is right that the TDI images shouldn't be regarded as quantitative at this stage - hopefully we'll be able to come up with a more quantitative variant in the future. That said, since you want to do these statistics, there are many ways to solve your problem - assuming I understand it correctly. </div>
<div><br></div><div>First off, I'm not sure why you want to use ImageJ at all for this. From what you say, it sounds like what you're doing is thresholding the CST TDI at a specific intensity and writing out a binary mask - this is what the <a href="http://www.brain.org.au/software/mrtrix/commands/threshold.html">MRtrix 'threshold' command</a> is for. Besides, I have a vague recollection that the last time someone here tried to use ImageJ, the image transform wasn't preserved so that the images were no longer aligned in real space - not a huge problem since mrstats doesn't use that information, and besides things may have changed since then anyway.</div>
<div><br></div><div>The next thing is as Luis suggested, there is a -template option for <a href="http://www.brain.org.au/software/mrtrix/commands/tracks2prob.html">tracks2prob</a> that will produce a TDI in exact alignment with the template image supplied. So you can generate your whole-brain TDI using:</div>
<div><br></div><blockquote style="margin:0 0 0 40px;border:none;padding:0px"><div>$ tracks2prob tracks.tck -vox X TDI.mif</div></blockquote><div><br></div><div>to get a high resolution image, and your ROI using:</div><div>
<br></div><blockquote style="margin:0 0 0 40px;border:none;padding:0px"><div>$ tracks2prob CST.tck -template TDI.mif TDI_CST.mif</div></blockquote><div><br></div><div>to guarantee alignment for use in mrstats, after having thresholded your image to create a binary mask:<br>
</div><div><br></div><blockquote style="margin:0 0 0 40px;border:none;padding:0px"><div>$ threshold TDI_CST.mif -abs N TDI_CST_mask.mif</div><div>$ mrstats TDI.mif -mask TDI_CST_mask.mif</div></blockquote><div><br></div>
<div>
Alternatively, you can regrid one image to match the other using the <a href="http://www.brain.org.au/software/mrtrix/commands/mrtransform.html">mrtransform command</a>, again using the -template option. That said, the tracks2prob route is the better option since it will avoid interpolation errors inherent in regridding.</div>
<div><br></div><div>Hope that helps.</div><div>Cheers,</div><div><br></div><div>Donald.</div><div><br></div><div><br></div><div><div class="gmail_quote">On 6 July 2012 07:59, Luis Concha <span dir="ltr"><<a href="mailto:lconcha@unam.mx" target="_blank">lconcha@unam.mx</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">You can use <div>tracks2prob -template TDI.mif CST.tck CST_p.mif<div><br></div><div>However, it puzzles me that you want to extract statistics from (presumably) ultra-high resolution images. FA values should, in my opinion, be extracted from the original FA maps. On the other hand, it has been suggested that track densities are not something that is a quantifiable parameter for between-group analyses. I am hoping this opinion has changed and/or that some new way of analysis of TDI is available. Perhaps others can pitch in.</div>
<div><br></div><div>Luis</div><div><br></div><div><br></div><div><br><div class="gmail_quote"><div><div class="h5">On Thu, Jul 5, 2012 at 4:39 PM, Kirles Bishay <span dir="ltr"><<a href="mailto:kirles@ualberta.ca" target="_blank">kirles@ualberta.ca</a>></span> wrote:<br>
</div></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div class="h5"><div>Hello,</div><div> </div><div>
I am currently trying to get statistics specifically on the corticospinal tract of my whole brain TDI images. So far, I have tried to create a mask of just the CST to apply to the whole brain image by using the tractography function in MRtrix, turning it into a TDI image, using ImageJ software to convert the TDI image into a mask and then applying the mask back in MRtrix to specify my ROI for the statistics function. While this works well with the TDI image of just the CST, it does not work with the whole brain image because the dimensions of the images are different. This is because the dimensions created with my ROI tractography differ from the dimensions of the whole brain tractography. I have tried to change the dimensions of my mask image using ImageJ, but I think this warps the mask inapropriately. Is there a way to control the output dimensions of the tractography images so that I can match the CST mask to my TDI image? Or do you have any other suggestions to get the data on the CST in my whole brain images that I have not thought of? </div>
<div> </div><div>Thank you,</div><div> </div><div>Kirles</div>
<br></div></div>_______________________________________________<br>
Mrtrix-discussion mailing list<br>
<a href="mailto:Mrtrix-discussion@www.nitrc.org" target="_blank">Mrtrix-discussion@www.nitrc.org</a><br>
<a href="http://www.nitrc.org/mailman/listinfo/mrtrix-discussion" target="_blank">http://www.nitrc.org/mailman/listinfo/mrtrix-discussion</a><br>
<br></blockquote></div><br></div></div>
<br>_______________________________________________<br>
Mrtrix-discussion mailing list<br>
<a href="mailto:Mrtrix-discussion@www.nitrc.org">Mrtrix-discussion@www.nitrc.org</a><br>
<a href="http://www.nitrc.org/mailman/listinfo/mrtrix-discussion" target="_blank">http://www.nitrc.org/mailman/listinfo/mrtrix-discussion</a><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br>Jacques-Donald Tournier (PhD)<br>Brain Research Institute, Melbourne, Australia<br>Tel: +61 (0)3 9035 7033<br>
</div>