<div dir="ltr">Hi all,<div><br></div><div>This is very interesting, the fact that you split the scan into 4. I think each scan will have different calibration though. Do you acquire a B0 for each of 15-orientation scans? So a question of principle is: doesn't this splitting affect the results at all?</div><div><br></div><div>Would be really nice if DWIs can be split like this.</div><div><br></div><div>Dorian</div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Sep 4, 2015 at 9:44 AM, Attye, Arnaud <span dir="ltr"><<a href="mailto:aattye@chu-grenoble.fr" target="_blank">aattye@chu-grenoble.fr</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear Mr Vanhoutte, Dear Dr Tournier,<br>
Firstly, I would acknowledge Mrtrix team and Dr Tournier for its wonderful software and this website.<br>
I’ve been working on peripheral nerves tractography for 3 years with Constrained Spherical Deconvolution model. We used the intraparotid facial nerve as a model to surgically validate CSD tractography data. I’m neuroradiologist so not expert in physics of MR diffusion imaging. However, I’m working with very good research engineers that help me applying CSD to peripheral nerves field. Our first tractography study will be soon published with methods details, but I'd be pleased to help Mr Vanhoutte with advices concerning facial nerve tractography:<br>
-The facial nerve is more difficult to track into internal auditive canal than in parotid area, mainly due to susceptibility distorsion and the vicinity with the vestibulocochlear nerve. Thus, diffusion acquisition requires a strong correction of artifacts. As described in the last article of Dr Tournier’s team (<a href="http://www.ncbi.nlm.nih.gov/pubmed/26163802" rel="noreferrer" target="_blank">http://www.ncbi.nlm.nih.gov/pubmed/26163802</a>), we use topup and eddy tools in FSL5 to correct B0 field inhomogeneities, eddy currents and inter-volume motion.<br>
-Track-density imaging is a wonderful tool before performing tractography with inclusion ROIs, especially among patients with IAC tumors. I personally generate TDI maps with a 0.3 mm isotropic voxel size.<br>
-As Dr Tournier said, Tractography accuracy is better with 60 directions or more than with 32 directions. However, we obtained good results in the parotid area with 32 directions and acquisition time is more compatible with a clinical setting.<br>
-An alternative way to limit movements’ artifacts is to split MR diffusion acquisition (eg 4 scans of 15 directions for a total of 60 directions).<br>
Best regards, Arnaud<br>
<br>
<br>
<br>
Dr Arnaud ATTYE<br>
<br>
Department of Neuroradiology<br>
<br>
Grenoble University Hospital - France<br>
<br>
<a href="http://www.neuroradiologie-grenoble.fr" rel="noreferrer" target="_blank">www.neuroradiologie-grenoble.fr</a><<a href="http://www.neuroradiologie-grenoble.fr/" rel="noreferrer" target="_blank">http://www.neuroradiologie-grenoble.fr/</a>><br>
_______________________________________________<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" rel="noreferrer" target="_blank">http://www.nitrc.org/mailman/listinfo/mrtrix-discussion</a><br>
</blockquote></div><br></div></div>