<div dir="ltr">For quantitative anatomical analysis, the manual delineation is treated as the gold standard, against which the accuracy of automated methods are measured. Therefore, there is no way to measure the accuracy of manual delineation. <div>
<br></div><div>For manual delineation, we usually do not ask accuracy (validity), but ask precision (reproducibility). Very often, reviewers ask data for test-retest or inter-rater precision data. Therefore, manual delineation protocols are important; you want to define anatomical areas where your structures of interest can be defined easily with clear boundaries. </div>
<div><br></div><div>In this respect, I think that the pre-processing of images with the affine transformation increases the precision. This is the "hybrid" approach described in Fig. 1A of the attached paper. This approach, at least, reduces the inter-rater variability to identify a slice level of interest because the delineation protocol can explicitly say, "choose Axial slice #90", for example.</div>
<div><br></div><div>Susumu</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Jan 30, 2014 at 7:11 AM, Shaimaa Abdelsattar <span dir="ltr"><<a href="mailto:shaimaa96@hotmail.com" target="_blank">shaimaa96@hotmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hello experts,<br>
I have one question regarding the degree of accuracy of my data if i performed affine and trilinear normalization of my subjects to a template and then i manually drew a single 2 d roi (5 voxel) on an anatomical structure e.g. left cerebral peduncle (trying to perform it in the exact level and location in each patient ) to compare the resulting data among subjects.<br>
Can this approach give a valid data to rely on in comparison<br>
<br>
Thank you for your concern<br>
Shaimaa<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>
</blockquote></div><br></div>