Blur Problems: Can the subject be included?
1: Sinus artifact
Sinuses below the temporal cortex (above the ears) and below the orbitofrontal cortex (behind the nose) are one of the great tragedies in the neuroimager’s professional life, as they obliterate two regions fundamentally involved in reward-processing, the engine of emotional life and the centerpiece of psychiatric research.
Intellectual honesty requires that before activation images are published indicating activity in one of these regions the researcher ensure that each individual subject’s functional scan is not plagued by sinus artifact.
The skills needed to do the following are covered later in the primer, but are given in detail here as from a procedural standpoint it is efficient to evaluate multiple sources of blur at the same time in order to decide whether or not to exclude a subject from analysis.
As shown in Figure 1,
1. Open FSL
2. Open FLIRT linear registration
3. Click on the Utils tab in the lower right hand corner, select ‘Apply FLIRT transform’, and a window saying ‘ApplyXFM’ appears.
4. For the first input line, Transformation Matrix, navigate the functional run that you want to check, and in the ‘reg’ directory, find the ‘example_func2standard.mat’ file and enter it. What this means is you are taking the functional run image and mapping it into standard space.
5. To prepare the input volume for line 2, go to a terminal window:
a. navigate to the folder where your raw data file of the functional run – the T2* image before preprocessing – is stored
b. type avwroi [enter] to see how the command wants information entered
c. assuming your raw data file is named r1.hdr.gz (run 1), enter the command ‘avwroi r1 r1first 0 1’ here r1 refers to your data file, r1first is the name you are giving to the first image of the functional run (because it is the first group hit by RF pulse, the first image has the highest contrast and is easiest for viewing), 0 means you want to start at the start of the run (FSL numbers the first volume 0 rather than 1, which you would think makes more sense) and 1 means you want to only go one image forward – eg a single slice. You could do more but it would be pointless.
d. now go back to ApplyXFM and enter the resultant image into line 2
6. For the Output size, the reference volume is the standard image (eg the MNI152 brain) which you can find a copy of in any ‘reg’ file in any functional run you’ve done on this data. In Figure 1B, you can see the ‘standard.hdr’ image has been loaded
7. For output, decide where you want the image to be. It is from this location that you will be pulling it into FSL, and you may want to look at this when your group results are done, so store it in a place where you can easily find it.
8. Name it ‘r1first2standard’ or some other canonical name that you will remember months from now.
9. Now you can import this image into FSL view, and overlay it over functional results. Ensure that your activations aren’t in a sinus artifact reason – if they are, they are due to mild head motion, not, obviously, to neural activity.
Figure 1: How to look at the first image from a functional scan in FSL
Larger voxels have a greater signal/noise ratio, but resolution is poorer. This matters if you are interested in small areas – you trade bad signal/noise for good resolution – but if you are interested in large areas, you trade resolution for signal/noise.
In theory, a higher field-strength magnet allows you to halve your voxel size (double your resolution) while maintaining your signal to noise ratio, which is why higher field strengths are preferred. However their magnification of sinus artifact is a major reason that researchers interested in the temporal and orbitofrontal cortices are somewhat wary of high field strengths.
3: Tesla Strength
Theoretically, noise remains essentially constant at any voxel size, while signal increases as a function of voxel size. This results in greater signal/noise ratios for larger voxels.
4: Checking for scanner drift
You can’t. This is not a problem because it is filtered out using a temporal filter; make sure you are satisfied with the one you used.
3: Checking for head motion
Deciding if head motion is too severe requires an eyeball test; there are no established standards for determining whether head motion, objectively, requires a subject to be thrown out of the study. This decision depends on several factors including:
1. How large the head movement was
2. When it occurred
3. Whether the subject returned to neutral position
How abrupt it was (gradual is better than abrupt).