Step 1: Slice Timing Correction (for event-related studies)

The statistical analysis you will perform later in SPM assumes that all voxels in the brain were acquired at the same moment, however the scanner acquires different slices continuously throughout the TR period. This step corrects for the different timing in acquisition of different slices within the same functional volume.

Slice timing is particularly important for event-related studies, but not so important for the analysis of blocked design studies.

(Type `help spm_slice_timing` for more in-depth information on this routine.)

Click on the “SLICE TIMING” button

This preprocessing step performs only within- session transformations. Because of this, you can perform in on one session at a time, or on all sessions at the same time. (In contrast, the next step, motion correction, performs both within-session and between-session transformations).

Click on Data. Create as many sessions in this category as you have sessions scanned.

Click on Number of slices. Enter the number of slices per volume collected.

Click on TR. Enter the TR in seconds

Click on TA. Enter the time between beginning of acquisition of the first slice and the beginning of acquisition of the last slice of each scan.

In typical (i.e. continuous) acquisition protocols
TA = (TR/#_slices)*(#_slices – 1)
The default value is calculated according to this formula.

If you used clustered (or “sparse”) acquisition sequence, the time to acquire all the slices may be significantly less than the TR. For example, it may take 1 second to acquire all slices within a functional volume, but the TR may be 2 seconds, in a clustered acquisition sequence.

For clustered (or “sparse”) acquisition protocols however, the gap between acquiring the last slice of a scan and the first slice of the next scan may be considerably longer than the gap between acquiring all sequential slices within the same scan. In such a case, you will need to know what proportion of the TR that was spent acquiring slices, and what proportion was spent in “silent mode”. In such cases, make sure to calculate and enter a TA different from the default values.

For example, your TR may be 2 seconds, of which the first 1-second interval was spent acquiring the slice, and the second 1-second interval was spent in “silent mode”. Let’s say you acquired 19 slices. Then, the time between beginning of acquisition of the first slice and the beginning of acquisition of the last slice within your TR will be:

TA = (time_spent_acquiring_slices / #_slices) * (#_slices – 1) = (1/19)*(19-1)=0.9474.

Click on Slice Order. Select the appropriate sequence.

”ascending (first slice=bottom)” means that the slices were prescribed and collected starting from the bottom of the brain and going to the top.

“descending (first slice=top)” means that the slices were prescribed and collected starting from the top of the brain and going to the bottom.

“interleaved (first slice=top)” is somewhat confusing, but it means that the first slice was collected in the middle of the brain, the second at the bottom, the third was collected right above the first, and the fourth right above the second, and so forth. For example, for 19 slices, the acquisition order would be: 10,19,9,18,8,17, and so on.

“interleaved (1 3..2 4..)” means that the slices where collected starting from the bottom of the brain, collecting all the odd numbered slices first (1,3,5…) and then collecting all the even number slices (2,4,6..)

“user specified” means that you can enter any order of acquisition by specifying the numbers so that 1 corresponds to the bottom slice and the number corresponding to the number of slices in your acquisition corresponds to the top slice.

For data collected at the UBC 3T Philips scanner, select the “interleaved (1 3..2 4..) option.

Click on Reference slice. Enter the slice you want to consider as a reference point. All other slices will be corrected to what they would have been if they were acquired when the reference slice was acquired. The default is the middle slice (although, please make sure the default value given is indeed the middle slice for the number of slices you have).

NB: The number you enter here is selected regardless of the acquisition sequence specified in ‘Select sequence type’ above. To decide what number you should enter here, disregard how your sequence was collected, and imagine a brain with slices numbered sequentially from bottom to top : 1 2 3 4 5 6 … up to your #_of_slices. Now select the number corresponding to the slice you would like to use as reference. The number you have now is correct if you are using the standard SPM release.

Deciding on reference slice
The logic behind selecting the middle slice as a refenrence point is that in this way the remaning slices will be closest to the reference slice and therefore any interpolation introduced by the correction procedure (and any potential errors due to it) would be minimized. Moreover, the further away a slice is from the reference slice, the larger the potential error is likely to be, and by selecting the middle slice in space as the reference slice, the bulk of the error would be pushed toward the top and bottom of the acquired volume, which usually are of least interest and, in the case of a full-brain acquisition, contain least gray matter. However, if a specific slice or a group of slices are of particular interest, it may be preferable to select the reference slice so that it reflects this.

SPM_fmri_defaults: The fMRI_T0 global variable specified in the spm_fmri_defaults.m file should be set to correspond to the reference slice selected above. fMRI_T and fMRI_T0 refer to the number of time-bins per TR. fMRI_T should be around 4-6 time bins per second, any higher temporal resolution will not gain any benefit for the statistics, and would results in unnecessarily increased time in statistics computation.