(function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); skip to primary navigationskip to content
 

Pre SPM Preprocessing

Your raw imaging folder will be in the WBIC directory:

/pvCache/pvuser/nmr

It will start with the WBIC subject number e.g. W010010 and will have a 3-character suffix ".ab1" to identify the session. You can identify the date of this session by typing:

ls -l W010010*

This raw data file is broken down into constituent fMRI runs using the pvconv routines (see http://pvconv.sourceforge.net). I run pvconv on the WBIC network to ensure the most up-to-date version, and because you have to copy your data from pvCache to your home directory anyway at some point. Create a directory in your home directory. e.g. home/lc260/study/010010 that you will pvconv the raw data into, and from here, type

pvconv.pl /pvCache/pvuser/nmr/W010010.abc -all -verbose -frecofix

The pvconv'd data can then be tarred up to transfer across to the BCNC network. From the directory /home/ab123/study type:

tar cvf 010010.tar 010010

Now from your BCNI account, type:

scp -rp gate1.wbic.cam.ac.uk:/data/ab123/study/010010.tar /work/imaging1/ab123/

And you can now untar the data:

tar xvf 010010.tar

This should create a series of 4d files in your subject directory, which will include the scans for your EPIs, phasemaps, structurals, templates etc - you can work out which is which from the MRI log that the WBIC radiographer should give you. (You can usually tell the EPI runs by typing ls -l and looking at the file sizes too). Within this main directory, you can now create a separate directory for each EPI run, for example, rev1, rev2, wm1, wm2. The ana4dto3d command will separate each 4d EPI run into a series of separate 3d files representing each individual scan. Before you do this, you need to find out how many dummy scans you used in your task, because ana4dto3d will discard these scans at this point. If you have 4 EPI sessions per subject, you will need to run ana4dto3d four times per subject. From the subject directory, type:

ana4dto3d -d 12 -m -v W010010_04.img /work/imaging1/luke/010010/rev1

where I have 12 dummy scans, and the fourth scan was the EPI run for task 'rev1'. The -m -v options create raw mean and raw variance scans, which you can now open in SPM, using the display function. The mean scan should look roughly brain shaped! Click around the image, looking for holes, tiger stripes etc. You should also look at the raw variance image: you might need to click Colour - Effects - Brighten a couple of times to make it out, but you should be able to make out the edge of brain where there is naturally high variance. This is a good thing. You don't want any lines going thru the brain, which could indicate slice dropout. For quality control and more on diagnostics, see tsdiffana.

Before starting your SPM preprocessing, there are a couple of things you might need to do: If you are planning to use Rhodri Cusack's EPI undistortion routines, you will need to copy the raw (ie. prior to pvconv, from the pvCache/pvuser/nmr folder) phasemap directories from the WBIC files, so in each subject directory, I create a directory 'rawphase' which contains two numbered directories for phasemaps a and c (eg. scans 13 and 14).

I also copy the structural SPGR scans for each subject into a single directory 'structurals'. If you want to send your subject a picture of their brain, you can open these in SPM or in MRIcro (type 'startmricro') and save a nice-looking view using a screen capture tool like Gadwin Printscreen. If you plan to use your structural scan in the normalisation process, you will need to trimand skull-strip your structural scan using bet. These routines run in SPM.