Dear all,

I have performed ROI to ROI and Seed to voxel analysis with 2 seed regions ( Right Subiculum, CA1 )  on 2 groups of subjects ( Healthy and patients with Vestibular failure ) , I am annexing some results for your reference ( results pertaining to Seed to voxel analysis )  .

From the Group comparison output contrasts [-1 1] ( Healthy < Vestibular failure ) I obtained 2 significant clusters with several suprathreshold voxels in Right Insular and Superior Parietal lobe ( see attachments ) . I would like to know how to interpret these results since the significant clusters cover just 12% of  Right insular , i would like to know if this result is quantifiable . Also is there any way i can perform Seed based correlation on two groups ( Healthy and Vestibular failure subject pool )  and generate Pair wise  correlation maps with my two seeds ( Right Subiculum and CA1 )  computing strength and sign of connectivity pattern in two groups and differentiate between them.

 

Thanks
Vasudev

Dear Vasudev

If I am interpreting correctly your results indicate that the average connectivity with Right Subiculum and CA1, and each of those two clusters (located around the right insula and superior parietal lobe) is higher in Healthy compared to Vestibular failure subjects. In order to explore separately the effect of each seed, I would suggest first to select on your current results-explorer window the 'export mask' button to save these two clusters into a new file. Then go back to your main CONN gui second-level results tab and select the same two seeds/sources as in this analysis but this time leave the default [1 0;0 1] between-sources contrast (instead of using your current [1 1] contrast). Last click on 'results explorer', and on the new window select 'import values', change the option that reads 'cluster of interest in current analysis' to 'other clusters of interest' and select the previously-saved mask file. After you do this you will end up with four new second-level covariates, each containing the connectivity between each of your two seeds (Right Subiculum and CA1) and each of your two clusters (approx right insula and left superior parietal lobe). You can then go to Tools.Calculator to display those values for each of your two groups in order to better characterize the between-group differences that you are observing. 

Hope this helps
Alfonso

Originally posted by Vasudev Devulapally:

Dear Alfonso,

Thanks for your response, The voxels in cluster ( Insula and Superior Parietal lobe ) has several anatomical regions , How can i possibly get the information pertaining to each anatomical region related the the voxels in the clusters (Insula and Superior Parietal lobe) . When you have suggested " average connectivity is higher in healthy controls compared to Vestibular failure group " , How should i quantify it , I mean how high is it compared to all other clusters.

I have done ROI-ROI analysis with Jeulich atlas where Insula and Superior Parietal Lobe were well defined ( see the results in attachment ) interestingly i did not notice any connectivity in Superior Parietal lobe region , but is see  Right hemisphere _Visual Cortex _V5 has higher connectivity w.r.t to R_subiculum and CA1. Why do i notice such differences between different methods of connectivity analysis 



Thanks
Vasudev