Some questions about Cl Report

Submitted by yaosir on

Dear experts,

I am using the Rest slice viewer to visualize the results of the 2-sample T-test. When I finally clicked the Cl. Report and wanted to see the remaining clusters in detail, I found that some of the peak of the clusters were located in white matter although the clusters contained both gray matter and whiter matter. Some people told me to ignore these clusters since they did not locate in grey matter, however, I still doubted if it was reasonable. My question is whether I should reports these clusters in the table of my paper and how to explain that their peaks were located in white matter.

Thank you in advance.

Best,

Vincent.

Hi Vincent,

Here is the first thing comes to my mind after reading your post: You need to figure out why there are significant activity in white matter first. 

If I am in your shoes, I will check if my registration (functional image to anatomical image, anatomical image to MNI or TLRC template) are correctly done. Also I will check if the correct transformation were applied to the final maps. Also, I will check if my smoothing kernel is too large when I was performing spatial smooth.

If registation are good and proper parameter for spatial smoothing are used, this activation in white matter should be very limitted.

 

And to answer your question of reporting results. I believe, "Partially reporting results" = "Not telling the truth". not acceptable. But this is just my 2 cents.

Yang 

Dear Yang,

Thank you for your rapid reply. I believe that the registation was correct in my case (I did it using DPARSF and checked by hand), and the smoothing kernel was 4mm. Therefore, this activation may not be due to these two reasons.

It should be noted that my colleagues told me that the case that peak located in white matter was not rare according to their experience. Additionally, I remembered that Dr. Yan also said in the multimedia course that we can not rely on report by software such as xjViewer because sometimes it was inaccurate. Is it the cause? Or is there any solution about the problem.

Finally, I agree with your point that "Partially reporting results" = "Not telling the truth", and I believe that even thought I do not report the results in the table, readers can also find it in the figure.

Best,

Vincent

Hi Vincent,

I agree finding differences within white matter is not rare. I think a practical way is reporting in the table as it is. However, in the discussion, you can just not pay attention to it if you didn't find a good way to interpret it -- focus on those on grey matter and discussable.

Best,

Chao-Gan

Forums
DPABI/DPABISurf/DPARSF