help
help > Temporal properties of networks dynamics
Mar 15, 2020 07:03 PM | Juan Pablo Princich - Neurociencias Hospital El Cruce, Buenos Aires
Temporal properties of networks dynamics
Dear all, thanks the authors and users for making this toolbox
available which i found priceless for non computer intensive users.
Now that we all need to stay at home for long, will appreciate advise for the analysis of resting fmri where i need to investigate differences in time dynamics of DMN connectivity between controls and patients.
Previous publications suggest that patients has significative differences in the variance, variability and frequency of DMN connectivity related to subcortical structures (Thalamus) and also with other networks.
Based on Conn user manual and forum posts i think the best approach would be #1. to perform 1st level analysis of DynIca circuits including DMN (Thalamus and other networks atlas) as an a priori ROI definition to isolate common circuits for patients and controls (GroupICA?)..
But it is not clear for me if then i may need to go for 2nd level analysis to compare temporal properties (frequency and variability) between patients-controls for each Dynamic Factor? It 'd be great to have a Results Explorer-like option to better investigate temporal characteristics or maybe a little more info in the resumed user manual chapter. Should i correct the significance results of temporal analysis for multiple comparisons? how stats are obtained here?
Is it possible with Conn to extract the time points were these differences (if presents) eventually occurs (approximate Dynamic number) on each individual patients? Will be helpful to use the onset time of these states transitions for further analysis.
Will also greatly appreciate your advice if a complementary DynIca 2nd level analysis of spatial properties may help to detect relevant nodes in the Dynamic Factor interaction that previously showed temporal differences? For instance , will this analysis identify which individual nodes inside the DMN account for major differences?
Sorry for the multiple questions.. will love to have your comments on alternative possibilities for this kind of analysis ( vox-to-voc ICA role?) and also recommendation for relevant ROI definition and pre processing better options than default.
Cheers from Buenos Aires
Juan P
Now that we all need to stay at home for long, will appreciate advise for the analysis of resting fmri where i need to investigate differences in time dynamics of DMN connectivity between controls and patients.
Previous publications suggest that patients has significative differences in the variance, variability and frequency of DMN connectivity related to subcortical structures (Thalamus) and also with other networks.
Based on Conn user manual and forum posts i think the best approach would be #1. to perform 1st level analysis of DynIca circuits including DMN (Thalamus and other networks atlas) as an a priori ROI definition to isolate common circuits for patients and controls (GroupICA?)..
But it is not clear for me if then i may need to go for 2nd level analysis to compare temporal properties (frequency and variability) between patients-controls for each Dynamic Factor? It 'd be great to have a Results Explorer-like option to better investigate temporal characteristics or maybe a little more info in the resumed user manual chapter. Should i correct the significance results of temporal analysis for multiple comparisons? how stats are obtained here?
Is it possible with Conn to extract the time points were these differences (if presents) eventually occurs (approximate Dynamic number) on each individual patients? Will be helpful to use the onset time of these states transitions for further analysis.
Will also greatly appreciate your advice if a complementary DynIca 2nd level analysis of spatial properties may help to detect relevant nodes in the Dynamic Factor interaction that previously showed temporal differences? For instance , will this analysis identify which individual nodes inside the DMN account for major differences?
Sorry for the multiple questions.. will love to have your comments on alternative possibilities for this kind of analysis ( vox-to-voc ICA role?) and also recommendation for relevant ROI definition and pre processing better options than default.
Cheers from Buenos Aires
Juan P
Threaded View
Title | Author | Date |
---|---|---|
Juan Pablo Princich | Mar 15, 2020 | |
Juan Pablo Princich | Mar 18, 2020 | |