Hi,

I am a researcher at UPenn and was wondering if there is a way to know a list of ROIs (names and their corresponding values in the image) that are present or expected in each of these images. I tried looking for it in the README.txt but what I found was a listing of ~80 ROIs that do not correspond with the images.

If you could please point me in the right direction, that'd be great!
Thanks,
Jimit

Hi Jimit,

The README.txt file has the proper numbers.  If you are more specific about which files you are talking about and how you are looking at the label numbers (i.e. what software), I'm sure we can figure out the problem.

Andy.

Hi Andy:

Thank you so much for your response! I am referring to the following README.txt that I downloaded along with the dataset. On line 340, it says value 84 refers to Corpus_Callosum, whereas there is no 84 in the IBSR_??_seg_ana.nii.gz files (I am using fslview to look at the images). However, the values for hippocampii and some other structures seem to be correct and coincide with the values in the image.

So I just wanted to make sure that I am looking at the right values within the image. A list of the unique values I found within the images IBSR_{01..18} are,

2 3 4 5 7 8 10 11 12 13 14 15 16 17 18 24 26 28 30 41 42 43 44 46 47 49 50 51 52 53 54 58 60 62

Does this mean that the images I downloaded simply do not have all ROIs (e.g. corpus callosum, left cerebral exterior, righ insula, left thalamus etc.), but only a subset of those in the list?

Please let me know if I am making a mistake somewhere or if I am looking in the wrong place! 
Thanks once again!
Jimit

Hi Both,

Was there a solution to the problem Jimit had?

I am facing the same issues i.e., number of labels in the Readme.txt is 84 and number in the actual segmentation is only 35. 

Akshay

Oops, it looks like I never responded to this. Sorry!

The README may list more labels than are actually present in the labeled scan.  

Andy.