Good morning everyone,

Our Center has been on a quest for fMRI.  We have a mobile scanner currently that is a 1.5T and is a Seimens Syngo.  I have been working with a company that supplies both software and hardware for fMRI as well as a Seimens rep.  From all of my research I was under the impression that fMRI is just simply software to interpret the images and that we would not need to do anything to the existing scanner or settings etc.  
I am recently told that is not the case and that we would need to update to a fixed scanner to be able to do fMRI sequences. 
I feel like I am getting the run around here and told many different things.  We are so anxious to get started running fMRI sequences and would like to do so without spending a million dollars.

Could someone please tell me if it is possible to get the software to interpret the imaging without this costly upgrade and if so how?  Would anyone be willing to help with this process?  We are in Flint, MI.

Thank you,
Rebecca

I do not know if I understand correctly your post, but what it seems to be is that you need a software to perform the image analysis of the fMRI studies. If so, you can work with FSL (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/), it is for free. You will need another tool to convert the dicom files to nifti files, you can use dcm2nii (https://www.nitrc.org/projects/dcm2nii/). There are free, you should be able to download and install them. You should check if you have the sequences in your MRI scanner to perform fMRI studies (ep2d_bold), SIEMENS's engineers may help you with that. FMRI studies are not that simple, in my opinion, there are a lot of things you should take into account in order to obtain a quality studio not only in the acquisition steps but also in the image processing. You should get someone to help you with that, in Miami there is someone who may go there (http://www.fmriconsulting.com/).

Regards,

Juan.

Hi Rebecca,

I think you may want to speak to a physicist about that. Do you have one at your institution? fMRI is both a different way to acquire the data and a way to analyze the data. fMRI acquisition takes a special sequence. Not all magnets are capable of performing the same sequences. Sometimes it's a hardware issue, and sometimes it's a software issue. It's possible that the mobile scanners don't have the flexibility for certain sequences. If you have never set up an fMRI experiment or analyzed the data, I would suggest forging a collaborative relationship with an expert who is interested in your specific topic.

Hope this helps,
-Marina

Hi Rebecca,

You are asking two separate questions. As mentioned in the replies you received to the question you asked several months ago, the most popular packages for analyzing fMRI data are FSL, SPM and AFNI.  Each of them is freely available.  Juan gave you the link for FSL.  You can easily find AFNI or SPM, if you Google them.

But as Marina says, not all scanners are the same.  fMRI is usually done using using a pulse sequence called Echo Planar Imaging (EPI), on the Siemens scanners I have used the protocol for EPI is called ep2d_bold (as I mentioned in your previous thread).  EPI is very demanding on the gradient system, but the gradient systems in current fixed scanners are usually up to the task.  I have never used a mobile scanner but it would not surprise me if gradient system in a mobile scanner is not capable of doing EPI.  Even if it is, I would be surprised if the stability were good enough to get useful fMRI results.  If Siemens is telling you that you can't do fMRI on a mobile scanner, I would believe them.

Sorry I can't be more encouraging,
Allen

Originally posted by Rebecca Aaron:

Hi,

I am new FSL user and installed on MacOS 10.12.4

I installed MRIcroGL and use dcm2niix to convert my Philips 3D T1 images for reconstruction using FSL.
one .nii.gz output with name starts with T1*** but show blank image in fslview

However, if I use the older version of dcm2nii I got 3 sets of *.nii.gz files. All these three images with file names starts with
20170216****
co20170216***
o20170216****

I can display all three files in fslview.

Which one I should use in FSL BET operation.

Thanks

Hi, I have been using the 20170216*** file that is just the original T1 converted to nifti format. The co20*** file is the same T1 but cropped and adjusted because the excess air surrounding the head as well as the neck have been removed. The o20*** is, I think, the re-oriented image. However, I do not know how it works but when I check this file it is always left-right flipped, so I do not use it at all.

Check this out http://www.cabiatl.com/mricro/mricron/dcm2nii.html, "Reorienting to canonical space" section.