help
help > RE: Correlation: patients' score & connectivity
Apr 2, 2015 12:04 AM | Alfonso Nieto-Castanon - Boston University
RE: Correlation: patients' score & connectivity
Dear Yong,
Yes, using the latest CONN release you may easily extract connectivity values from your suprathreshold clusters (simply click on the 'Import values' button on the second-level results explorer). That will extract the average connectivity values for each subject (between your seed ROI of thalamus L and your suprathreshold cluster -136 volxes-) into a new second-level covariate variable. You can then go to 'Tools.Calculator' to display those values and their association with patient scores (e.g. select the new "connectivity with cluster ..." variable in the 'Outcome variables' list, and then select the 'patient' and 'score' variables in the 'Predictor variables' list and enter a [0 1] between-subjects contrast there)
Hope this helps
Alfonso
Originally posted by Yong Li:
Yes, using the latest CONN release you may easily extract connectivity values from your suprathreshold clusters (simply click on the 'Import values' button on the second-level results explorer). That will extract the average connectivity values for each subject (between your seed ROI of thalamus L and your suprathreshold cluster -136 volxes-) into a new second-level covariate variable. You can then go to 'Tools.Calculator' to display those values and their association with patient scores (e.g. select the new "connectivity with cluster ..." variable in the 'Outcome variables' list, and then select the 'patient' and 'score' variables in the 'Predictor variables' list and enter a [0 1] between-subjects contrast there)
Hope this helps
Alfonso
Originally posted by Yong Li:
Dear Alfonso,
I have performed an analysis to study correlation between patients' clinical scores and their connectivity by selecting 'Patient' and 'Score' and entering contrast [0 1]. As results, for example, I found a significantly correlated ROI-voxel connectivity cluster (size 136 voxels) in Premotor L with the peak coordinate [x, y, z] from a seed ROI of thalamus L. Is there an easier way e.g., a newer version of CONN, to extract connectivity results at the peak coordinate per patient so as to produce plots for publication (see example plot in our last paper, Yong_correlation.jpg)? I did it manually with SPM last time.
Thank you again!
Kind regards,
Yong
I have performed an analysis to study correlation between patients' clinical scores and their connectivity by selecting 'Patient' and 'Score' and entering contrast [0 1]. As results, for example, I found a significantly correlated ROI-voxel connectivity cluster (size 136 voxels) in Premotor L with the peak coordinate [x, y, z] from a seed ROI of thalamus L. Is there an easier way e.g., a newer version of CONN, to extract connectivity results at the peak coordinate per patient so as to produce plots for publication (see example plot in our last paper, Yong_correlation.jpg)? I did it manually with SPM last time.
Thank you again!
Kind regards,
Yong
Threaded View
| Title | Author | Date |
|---|---|---|
| Yong Li | Apr 1, 2015 | |
| Alfonso Nieto-Castanon | Apr 2, 2015 | |
| Yong Li | Apr 2, 2015 | |