External Email - Use Caution
Dear FreeSurfer community, I am running retinotopy analysis with FSFAST using:
- FreeSurfer version: FSBUILD freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0-2beb96c - on Linux 4.15.0-55-generic
I have only the polar angle condition, in two runs, one with positive and second with negative direction. To process the data, I followed the steps listed here: https://kylemathewson.com/retinotopy-in-freesurfercnl-protocol/#FUNC For the analysis to run, I created "dummy" eccentricity folders, as advised here: https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg48640.html
The pipeline seem to have worked (no error messages, no warnings, I enclose the analysis.info http://mkanalysis.info/ and the selxavg.log files).
I am now trying to visualize the data to check the results. In the tutorial I followed, it is recommended to view the real.nii.gz and imag.nii.gz files. You will find the screen shots of those files visualized in freeview attached.
To my eyes, none of those images look like polar angle retinotopy maps. I then visualized the angle.masked file (if I understand correctly, it is the angle map masked with p=0.01 significance), you can find the screen shot attached. It also doesn't look right, more importantly - it doesn't seem to be masked by significance.
I have then followed the visualization instructions from the official FreeSurfer wiki, i.e: tksurfer-sess -a rtopy.self.?h -s sessid -map angle.masked I attach the screen shot (tk_surfer_file.png). Frankly, I'm not sure what I'm looking at there :)
Looking for documentation that would describe retinotopy output files, I found the selfreq.pdf file, a documentation explaining *programs for computing the amplitude and phase of the hemodynamic response at a certain frequency.* In this file I found a c shell script to visualize the processed data:
#!/bin/csh -f
set name = subject # put the subject’s name here
set hemi = lh
setenv eccendir eccen
setenv polardir polar
4
#setenv patch occip.patch.flat
setenv revpolarflag 1
setenv rgbname eccenthinmot
setenv fthresh 0.4
setenv fslope 1.3
setenv fmid 0.8
setenv angle_offset 0.77
setenv noexit
tksurfer -$name $hemi
I adjusted this script to my environment and run it on my machine. It doesn't work because it's missing the following input files:
sigf-lh.w
sig2-lh.w
sig3-lh.w
Indeed, those files do not exist in my folders. Does it mean that I made a mistake somewhere in the pipeline? Should those files be in my folders?
To sum up, these are my questions: 1) Do the enclosed maps seem alright to you? 2) If not, do you have any idea at which processing stage there was an error? Or is it just a visualisation issue? 3) If this is a visualization error, how should I visualize the angle maps in freeview? Specifically, what output files should I use for this visualization? More specifically, what commands should I use to visualize these output files? For your information, these are my output files in the retino.lh/polar folder:
angle.masked.nii.gz cesvar.nii.gz fsig.nii.gz mc.sig.dat angle.nii.gz cesvarpct.nii.gz imag.nii.gz minsig.nii.gz cesmag.nii.gz cwmap-imag.nii.gz iminsig.nii.gz real.nii.gz cesmagpct.nii.gz cwmap-real.nii.gz mag.nii.gz sig.nii.gz ces.nii.gz efficiency.dat mc.ces.dat z.nii.gz cespct.nii.gz F.nii.gz mc.minsig.dat
4) what do the imag.nii and real.nii maps correspond to? I believe they represent the real and imaginary component of the Fourier transform, but how do I interpret it in terms of retinotopy results?
I would be very grateful for any help.
Take care, Kasia
