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
Well, for measuring temporal AC, censoring would certainly not be appropriate, either with zeros or by collapsing.
High motion will be destructive to this in one way or another, for all of these cases.
So given the choice between no censoring and interpolated censoring, it is difficult. High motion has drawbacks for each. If no censoring is done, that will probably drive down the AC values.
by
rick reynolds
-
AFNI Message Board
Hi Philipp,
It would certainly be possible to add such an option. But I worry about the use. Not only will interpolation (inappropriately) add to the autocorrelation, but note that autocorrelation itself is expected in what is considered akin to a BOLD time series (rest after nuisance projection). How are you planning to apply the autocorrelation measures or results?
- rick
by
rick reynolds
-
AFNI Message Board
I would expect the autocorrelation to be highly inflated by any termporal interpolation.
Assuming this is task data, performing a regression with 3dREMLfit should account for an ARMA(1,1) noise model, rather than using a simple least squares fit method, as would be done by 3dDeconvolve or 3dTproject.
Presumably this is not rest data.
- rick
by
rick reynolds
-
AFNI Message Board
That looks okay, though I am curious why you would want to do this. What will be the next step(s) with the result?
- rick
by
rick reynolds
-
AFNI Message Board
That is great and somewhat expected. To be sure, is this a system that is maintained by separate administrators? If so, you will probably need to ask them to install any needed libraries. The install instructions for CentOS 7 include these packages (yum install commands):
sudo yum install epel-release
sudo yum install gcc motif gsl libXpm
sudo yum install mesa-libGLw mesa-libGLU libXi gli
by
rick reynolds
-
AFNI Message Board
Yes, this output is very helpful. You are working on a CentOS 7 system there. We actually have binaries for that particular system (which might possibly need libraries, we can see). But change to those binaries and please post the afni_system_check.py output again. So run this first:
@update.afni.binaries -package linux_centos_7_64
Then please post the updated system_check output.
Tha
by
rick reynolds
-
AFNI Message Board
Hi Philipp,
What are the current values in the ROIs, and what would you like the values in the output to be? The 2 expected cases are either 0/1 or 0/VALUE (for input, say), where VALUE is perhaps an ROI index. Then the output might be either 0/1 or 0,...,MAX_VALUE, where MAX_VALUE would either be the number of ROIs or perhaps even some larger number, where ROI IDs are not necessarily 1,2,3,
by
rick reynolds
-
AFNI Message Board
Hi Angelika,
EDIT: Oh my, Paul has gotten me back... maybe this post is moot.
Different software has created NIFTI datasets with different intentions in the sform/qform_code fields, making a value of 2 there ambiguous. It is generally applied incorrectly even.
To be sure, what is the output of:
nifti_tool -infiles 1002002_V001_DMSRun1.nii -disp_hdr -field sform_code -field qform_code
by
rick reynolds
-
AFNI Message Board
Thanks for the script and timing files! I have pushed a change for this which should be built this weekend.
Unfortunately, the comment about detecting global times comes after the damage was done from not knowing, possibly leaving you with far too few regressors.
The code was updated to make global timing be the default for the assessment of the number of trials, and to give a warning if I
by
rick reynolds
-
AFNI Message Board
Hi Matt,
Indeed, there is some problem in 3dDeconvolve, possibly with the combination of global timing and IM. The problem seems to happen before 3dDeconvolve even detects that the timing is global. Based on that, you should be able to fix the problem simply by specifying the timing type BEFORE any timing files are given:
-global_times
So add that option before any -stim_times_IM option
by
rick reynolds
-
AFNI Message Board
Hi Matt,
Would you mind posting a sample of one of the timing files that is not giving you the number of betas that you expect? Or, you could mail me the timing file along with any X.xmat.1D file (or a nocensor version).
Are you using afni_proc.py? If so, try adding -regress_compute_fitts. That will use 3dcalc to compute the fitts, rather than having 3dDeconvolve (use the extra RAM to co
by
rick reynolds
-
AFNI Message Board
The 3dresample step is not necessary, but it makes this more of a predictable change. If the values on the original grid are shuffled around such that it is as if the original scanner orientation was RAI, then the sphinx position correction should always be the same 3drefit command, which Paul is saying should use -orient LIP. Without the resample step, Paul would not know what to suggest for 3
by
rick reynolds
-
AFNI Message Board
Hi Silvia,
It would be very fast to rerun the same 3dttest++ command without -ClustSim. In that case, the output would include the t-stat instead of the z-score.
And it would probably run in just a few seconds. So give it a new -prefix option and remove -ClustSim.
You might also want to use 3dbucket (for example) to extract the t-stat into a separate NIFTI dataset.
- rick
by
rick reynolds
-
AFNI Message Board
Since the time series was thresholded into a binary spike time series in step 1, the ROI average is indeed the fraction of voxels that are spikes at each time point. It is not the average spikiness, but the average of the spiked voxels.
For example, if at some time point 40% of the ROI voxels are 1.5 and 60% are 3, the binarized time series will have 40% of the voxels at 0 and 60% at 1, since
by
rick reynolds
-
AFNI Message Board
During a hackathon a bit ago, Bob Cox was working with the ANTs folks about making these transformations play together better, but I don't know how far that got. For a simple affine, it shouldn't be toooo complicated. But certainly, don't make your life more difficult than it needs to be.
- rick
by
rick reynolds
-
AFNI Message Board
This may be a slightly messier issue than you are picturing. I am not confident about warping tensor files, since the warps would not modify the rotations in the tensors themselves, and it seems like that would be important. The warp would merely move the tensors, not re-evaluate them.
Ignoring that, note that most AFNI programs do not even handle 64-bit float data. NIFTI images will be con
by
rick reynolds
-
AFNI Message Board
I think 3dANOVA3 requires a complete input table. So with 2 conditions x 2 time points x 14 subjects, there should be 56 = 2x2x14 -dset entries, where each subject has 4 entries. Or are the conditions separate between 2 groups of 7 subjects? If so, then perhaps clevels should be 7.
- rick
by
rick reynolds
-
AFNI Message Board
Hi Carolin,
I am pretty sure Paul had meant running 3dToutcount on the original time series, not the 3dDespike -ssave output. But let me just go from what you have said you want for now. You wanted to:
- run 3dDespike -ssave
- calculate the fraction of voxels in your ROI where s exceeds 2.5, say
- count the number of time points where the fraction exceeds a a threshold
So assuming
by
rick reynolds
-
AFNI Message Board
Try the following command. If it returns nothing, then you do not have that pnmtopng program installed. It comes with the netpbm package.
which pnmtopng
I should add a test for one of those programs to afni_system_check.py...
- rick
by
rick reynolds
-
AFNI Message Board
Oh, and -jpg works for me. This is specific to -png.
- rick
by
rick reynolds
-
AFNI Message Board
Hmmm, I get the same bad behavior. I cannot produce a png file (even with Paul's command). Maybe that means there is some shell program that is needed that I do not have installed.
Will investigate...
- rick
by
rick reynolds
-
AFNI Message Board
Hi Catherine,
Thank you for the reminder.
I don't recall what I was thinking back then (big surprise), but there are some notable points from that mmms output.
There are 4 matrix columns that show min = max = stdev = 0, and 2 that are very close to that. What do these 6 columns look like?
I suppose 43, 44, 49 and 50 are all zero. But 42 and 48 are surely close to that. Any all-z
by
rick reynolds
-
AFNI Message Board
Hi Clement,
Note that the default for 3dttest++ is -center DIFF, though I guess you should specify that anyway.
In the case of -center DIFF, each group will have any covariate means removed separately. So here the age covariates should be orthogonal to the grouping terms. At least that would apply to 3dttest++...
Does that seem reasonable?
- rick
by
rick reynolds
-
AFNI Message Board
Hi Brooke,
If you are happy with dcm2niix, that is great. But if you would like to investigate the stretching via Dimon, it is probably necessary for me to get a hold of some of the data. In that case, we should move the discussion to email. Just let me know if you would like to proceed with that, it is totally up to you. No pressure either way.
Thanks,
- rick
by
rick reynolds
-
AFNI Message Board
Glad you have it working.
Stretching sounds like a problem with the dimensions within the DICOM files.
Would you please do one more test by adding -use_last_elem to the Dimon command?
Thanks,
- rick
by
rick reynolds
-
AFNI Message Board
Yes, the issue in the first point would arise because of the separate motion regression.
It would be more correct to orthogonalize the regressors of interest with respect to motion and polort, whatever is in the initial regression.
But really, what is the benefit of running 2 regression steps (projecting nuisance before computing betas)?
- rick
by
rick reynolds
-
AFNI Message Board
Hi Sanjana,
Sure, modeling that in just the last 4 runs makes sense. I had initially gathered that all runs were in a single model.
Regarding running an initial regression before running a secondary regression on the residuals, there are 2 distinct issues with it.
1. Assuming the betas in the first regression are not important (such as for motion in this case), a secondary regression is
by
rick reynolds
-
AFNI Message Board
For this you might run Dimon (or dcm2niix_afni) first and just let them assume the orientation is correct.
Then run 3drefit -orient afterwards to set the orientation to what it should be.
Note that right vs left might be hard to verify. Hopefully you have a method to be sure of which side is which.
Does that seem reasonable?
- rick
by
rick reynolds
-
AFNI Message Board
Maybe I am not quite following what the input is. Is this just the regression of phase 2, the last 4 runs? And in phase 2, you want the model b1*S1 + b2*S2 + b3*S1*S2? Or is that model supposed to span all 8 runs? That aspect is not quite clear.
In general, it is more appropriate to run a single complete regression than to project nuisance before a secondary one.
But ignoring that, I ju
by
rick reynolds
-
AFNI Message Board
Hello,
That looks okay, depending on how s1*s2 was constructed.
Since the phase of the S1 and S2 events differs from that of S1xS2 (is that the p2 file?), presumably the S1.1D and S2.1D files are zero where S1xS2.1D is non-zero, and vice versa. So if those were multiplied, the result would be all zero. Is that the case? I just want to be sure the regressors are temporally offset, so that
by
rick reynolds
-
AFNI Message Board