Show all posts by user
Dear AFNI users-
We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:
https://discuss.afni.nimh.nih.gov
Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.
The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.
Sincerely,
AFNI HQ
History of AFNI updates
Page 2 of 3
Pages: 123
Results 31 - 60 of 72
Sure, please let me know when your iron has cooled down ...
Thanks!
by
Zhihao_Li
-
AFNI Message Board
Thank you Gang for your prompt response!
My scenario is very simple:
resting-state fMRI
2 groups, control and patient
anatomically derived 200 ROIs
200x200 functional connectivity matrix obtained individually (so 19900 links for each pair of ROI)
(1) link-wise group comparison between controls and patients; or group t-test performed for each one of the 19900 links
(2) link-wise correlat
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI experts,
I am wondering if AFNI has an implementation of this "network-based statistic". It is similar to AlphaSim, but applied to "links" rather than voxels.
by
Zhihao_Li
-
AFNI Message Board
Thank you Daniel!
@radial_correlate does correlation in temporal domain while I want to do correlation in spatial domain. Or, I want to see if two spatial patterns (intensity distribution in space) are correlated.
I figure that I can split my process into 2 steps: (1) to use "3dLocalstat -list" to convert spatial patters into temporal patterns; and then (2) to use 3dTcorrelate to ge
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI experts,
I am wondering if there is a convenient way of doing spatial correlation (3ddot feature) in a moving "search light" spot (3dLocalstat feature). Namely, to make the "-mask" option of 3ddot spatially dynamic. I know the dumbest way would be to create a bunch of masks and loop through them with a bunch of 3ddot, but it is always the best to consult AFNI geniuses
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI experts,
I was given some ANALYZE data (hdr/img files) and need to convert them into AFNI format for further analysis.
Howver, even when trying to convert only 2 volumes, "to3d" complains about the slice number which I don't understand. Please see the following details:
3dinfo file1.hdr
R-to-L extent: -83.772 -to- 105.228 -step- 3.000 mm [ 64 voxels]
A-to-P
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI experts,
I am not quite sure how the "maximum pairwise displacement" is calculated in "1d_tool.py", please see if my understanding is correct. Suppose we have only 3 time points and the 6 head motion parameters at each time points are (a1, b1, c1, d1, e1, f1), (a2, b2, c2, d2, e2, f2), (a3, b3, c3, d3, e3, f3). So the "maximum pairwise displacement" is calcu
by
Zhihao_Li
-
AFNI Message Board
afni -ver
Precompiled binary linux_openmp_64: Nov 23 2016 (Version AFNI_16.3.13)
The option of "-Clustsim" of 3dttest++ does work after saying "Using 15 OpenMP threads".
However, my 3dClustSim still hangs exactly there. Weird.
by
Zhihao_Li
-
AFNI Message Board
Thank you Rick, please the output of afni_system_check.py:
system with 3dClustSim hangs:
-------------------------------- general ---------------------------------
architecture: 64bit ELF
system: Linux
release: 2.6.32-573.el6.x86_64
version: #1 SMP Wed Jul 1 18:23:37 EDT 2015
distribution: Red Hat Enterprise Linux Server 6.7 Santiag
by
Zhihao_Li
-
AFNI Message Board
I am showing you the output of "lscpu", but please let me know if you need any specific information.
The new version of 3dClustSim is running on this system:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 64
On-line CPU(s) list: 0-63
Thread(s) per core: 2
Core(s) per socket: 8
Socket(s):
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI experts,
I have 2 versions of 3dClustSim, one goes with AFNI_16.2.07 (older, Aug.2,16) and the other goes with AFNI_16.2.19 (newer, Sep.28,16).
I was trying to run it with "-acf" and with identical inputs. However, the old version works but the new version hangs.
The initial outputs are the same for the 2 versions except after "Startup clock time", the old version b
by
Zhihao_Li
-
AFNI Message Board
Hi Gang,
I recall that you mentioned earlier this year in an AFNI workshop that a new feature for group analysis of inter-subject correlation (results of 3dTcorrelation) would be available. I would like to check with you and see if this feature has already been distributed.
Specifically, I have several groups of subject (say patients and controls) scanned for many times (say visit_1 and visi
by
Zhihao_Li
-
AFNI Message Board
Hi Gang,
My AFNI has just been updated earlier this month for the newly added feature of autocorrelation modeling in 3dClustSim. So I assume the version is fine:
Precompiled binary linux_xorg7_64: Aug 2 2016 (Version AFNI_16.2.07)
Please find below part of the data table (lengthy data path omitted):
----------------------------------------------------------------------------------
Subj GR
by
Zhihao_Li
-
AFNI Message Board
Hi Gang,
I want to run a group analysis with 1 between subject factor: GROUP (Main, Control), and 2 with subject factors: TASK (Video, Language), VISIT (1st, 2nd, 3rd). There are missing data for some within-subject factors.
I was trying the following "3dMVM" and "3dLME" commands but got similar error information that I don't understand. Could you please help? Thank you
by
Zhihao_Li
-
AFNI Message Board
Thank you Rick, but "-layout" doesn't work with an error message of "LAYOUT not found in layout file a.script".
For the same layout, if I save it with a name of "a.layout", the content is:
A geom=+2+23
...
But if I save it with a name of "a.script", the content is:
OPEN_WINDOW A geom=+2+23
...
So the "-layout" only reads in thin
by
Zhihao_Li
-
AFNI Message Board
Thank you Paul for this prompt response!
Actually what I meant to ask was whether there is a SINGLE option that can read in a lot of command lines saved in a script file. Including many specifications of window positions, my script file has tens of lines but I want to avoid typing tens of "-com".
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI gurus,
On AFNI startup, I wanna simultaneously open 2 controllers with different underlay datasets displayed. I did so manually and saved the layout as a "script" file. Suppose the file name of this script is "a.script", how can I execute it on AFNI startup? I have tried "afni -com a.script", but it doesn't work with an error message of "ERROR: C
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI Gurus,
If both "-demean" and "-derivative" are used in an "1d_tool.py" command, what's the sequential order of these two operations?
by
Zhihao_Li
-
AFNI Message Board
I got the same error message with 3dfim+ and I happened to see this post.
In my case, after comparing datasets that "3dfim+" complains or not, I noticed that it is actually complaining that the input dataset does not have a time axis (a bucket instead of a 4D). So I provided (arbitrary) TR information with 3drefit and it works.
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI experts,
I am performing a voxel-wise correlation between values of activation beta and a behavioral measurement.
I usually do this with 3dfim++, which gives a sub-dataset of "Correlation". For visualization of this result, I put the "Correlation" as the threshold control (Thr), then I can move the slide bar up and down to see correlation maps at different p th
by
Zhihao_Li
-
AFNI Message Board
Thanks Gang,
I tried 3dLME, but got the same error of "Model test failed".
3dLME \
-model 'Dffclty'
-num_glt 4 \
-gltLabel 1 "d1-d3" -gltCode 1 'Dffclty : 1*d1 -1*d3' \
-gltLabel 2 "d2-d4" -gltCode 2 'Dffclty : 1*d2 -1*d4' \
-gltLabel 3 "d3-d5" -gltCode 3 'Dffclty : 1*d3 -1*d5' \
-gltLabel 4 "d4-d6&quo
by
Zhihao_Li
-
AFNI Message Board
Hi Gang,
I have a fMRI task with 6 levels of difficulty (d1, d2, ..., d6) but each subject only performed the task with 4 difficulty levels. For example, sbj1 performed and sbj2 performed . In this situation, can I run 3dMVM to examine the difficulty effect? I have tried the following but it says "Model test failed". Thank you so much!
3dMVM \
-bsVars 1 \
-wsVars Dffclty -wsE2 \
by
Zhihao_Li
-
AFNI Message Board
Thank you Rick for your input!
Like you have pointed out, I have been using 3dmaskdump and 3dUndump for this purpose.
I understand your concern of high correlation values. In my analysis, I use a spatial mask of 19 voxels over a "3d+t" of 463 time points, which I think is OK. The issue I have here is the computational load since I need to move the mask in the brain as a "search
by
Zhihao_Li
-
AFNI Message Board
I think that "3dTcorrelate" provides the most similiar function for this purpose. However, for canonical correlation, the calculation should not be between corresponding voxels but between corresponding ROIs of "xset" and "yset". Namely, I want to move a spherical ROI (user defined radius) in the brain with the output reporting voxel-wise canonical correlations. I wa
by
Zhihao_Li
-
AFNI Message Board
Hi Rick, I would still want beta weights over space.
The idea comes from this paper and I would like to try it with my data using AFNI.
Would you like to take a look at the method and recommend on AFNI tools that I can use?
Thank you so much for your help!
by
Zhihao_Li
-
AFNI Message Board
Hi AFNI experts,
I am wondering if there are any existing AFNI tools that I can use for analysis of canonical correlation.
Namely, given 2 sets of time series (by specifying a mask on a 3d+t and another mask on another 3d+t), X = (X1, ..., Xn) and Y = (Y1, ..., Ym), to find linear combinations of A=a1*X1+a2*X2+...+an*Xn and B=b1*Y1+b2*Y2+...+bm*Ym, which can maximize the A-B correlation.
I g
by
Zhihao_Li
-
AFNI Message Board
But it's not the statistical value (t in this case, which looks fine) that I am concerning. In AFNI's dataset of intensity/statistics, it is the "intensity" value that I am talking about. I am not quite sure how this "intensity" value is calculated in 3dMVM, but based on regular ANOVA, with individual inputs falling in the range of (-1.5, 1.5), with the output "
by
Zhihao_Li
-
AFNI Message Board
Hi Gang,
My seeding time course was an average of many voxels so no individual voxel has a correlation coeffient close to 1.
I have checked the range of my input z-scores with 3dinfo, actually none of them have exceeded 1.5 on fisher's z. So in this situation, can you think of any possible mistakes that I might have made?
by
Zhihao_Li
-
AFNI Message Board
Hi Gang,
I was running 3dMVM for group analysis of functional connectivity with inputs being fisher's z-scores.
My model is very simple with "bsVars" being Group (patient vs. control) and "wsVars" being Time (first visit vs. second visit).
I didn't do anything fancy but just demanded outputs of the 4 conditions of patient_first, patient_second, control_first, and
by
Zhihao_Li
-
AFNI Message Board
Page 2 of 3
Pages: 123