I see. Thanks a lot, Donald
Best, Feng
发件人："MCLAREN, Donald" mclaren.donald@gmail.com 发送时间：2015-10-07 11:01 主题：Re: [Freesurfer] Questions about the correction methods in Surfstat 收件人："Freesurfer support list"freesurfer@nmr.mgh.harvard.edu 抄送：
The output figure is attached. In the colorbar of the figure, left represents cluster corrected p value and right one represents the vertex >corrected p value, right?
Right.
So, the warm color region in the figure can be survived by the RFT vertex correction, and the cool color region in the figure can be >survived by the RFT cluster correction, right?
Right. RFT vertex correction is not a correct definition, it is a RFT correction based on the maximum peak in contrast with the larger cluster extent approach.
If so, the regions with either warm or cool color region that can be survived after correction. When I report the results, I can report the >regions with warm color (vertex level correction) or report the regions with cool color (cluster level correction), right?
Both thresholdings can be reported but I prefer peak thresholds (warm colors) due to the lack of spatial specificity with clustering thresholds.
In addtion, how can I understand the region overlapped by the warm and cool color (like he postier cingulate cortex), can >simultaneously pass double correction of vertex level and cluster level?
It is not a double corrections. The figure simply shows two different RFT thresholding approaches.
The final question is about the RFT cluster correction method, in the SPM or other software (like REST). Generally, on the cluster level correction, you should be based on a voxel (vertex) uncorrected height threshold of e.g. p<0.001 combined with a RFT-correlated cluster threshould of e.g., p<0.05, right? What is the detailed correction procedure in the Surfstat software?
Exactly the same, p=0.001 for defining supra-threshold clusters and p=0.05 for RFT correction.
Regards, Felix.
Best Regards, Donald McLaren, PhD
On Sun, Sep 13, 2015 at 9:36 AM, chenhf_uestc chenhf_uestc@163.com wrote:
Dear Surfstat Users,
I have a question about the RFT correction method in the Surfstat(http://www.math.mcgill.ca/keith/surfstat/). In the online manual, the command to perform RFT correction is as follows: [ pval, peak, clus ] = SurfStatP( slm, mask );
pval.P contains P-values for peaks, and pval.C contains P-values for clusters. This special structure is recognised by SurfStatView which draws the figure in a special way:
SurfStatView( pval, avsurf, 'Males-females removing age' );
The output figure is attached. In the colorbar of the figure, left represents cluster corrected p value and right one represents the vertex corrected p value, right? So, the warm color region in the figure can be survived by the RFT vertex correction, and the cool color region in the figure can be survived by the RFT cluster correction, right? If so, the regions with either warm or cool color region that can be survived after correction. When I report the results, I can report the regions with warm color (vertex level correction) or report the regions with cool color (cluster level correction), right?
In addtion, how can I understand the region overlapped by the warm and cool color (like he postier cingulate cortex), can simultaneously pass double correction of vertex level and cluster level?
The final question is about the RFT cluster correction method, in the SPM or other software (like REST). Generally, on the cluster level correction, you should be based on a voxel (vertex) uncorrected height threshold of e.g. p<0.001 combined with a RFT-correlated cluster threshould of e.g., p<0.05, right? What is the detailed correction procedure in the Surfstat software?
Any help would be greatly appreciate.
Best, Feng
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.