JISCMail - FSL Archives

Thanks for your reply!

(1) As for the timing value, your answer is a little different from your previous one (https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1703&L=FSL&P=R87264&1=FSL&9=A&J=on&d=No+Match%3BMatch%3BMatches&z=4).

My apologies - I made a mistake in the email I sent you as I was rushing. You are right that there should not be both +0.5 and -0.5. So the numbers should have been 0.4 0.2 0 -0.2 -0.4 etc, as I had put in this earlier reply on the list.

(2) For a real example, in which my BOLD image has 28 slices and (odd slice first) interleaved acquisition, I perform slice timing correction with two different ways (--ocustom & --tcustom), specifically

(2.1) slice order file (one number per row): 2 4 6 8 10 12 14 16 18 20 22 24 26 28 1 3 5 7 9 11 13 15 17 19 21 23 25 27

(2.2) slice timing file (one number per row, slice 28 is time 0): -1/28 13/28 -2/28 12/28 -3/28 11/28 -4/28 10/28 -5/28 9/28 -6/28 8/28 -7/28 7/28 -8/28 6/28 -9/28 5/28 -10/28 4/28 -11/28 3/28 -12/28 2/28 -13/28 1/28 -14/28 0

and I find the results are highly similar but different (I also tried to set the slice 1 as time 0 and modify the slice timing file accordingly, but it is still different from the slice-order-file corrected one ). I believe they should be completely the same if the two ways are equivalent. Any suggestions? Thank you very much!

This looks fine to me. When you say they are "highly similar" - are we talking about differences that are insignificant and potentially due to rounding errors (e.g., differences in intensities that are well under the noise floor)? Or are they bigger differences? It maybe that internally it is choosing to have the t=0 point in between acquisitions, rather than either slice 28 or slice 1, but I'd have to check in the code to verify this.

All the best,

Mark

Yang Hu

At 2018-01-06 01:13:14, "Mark Jenkinson" <[log in to unmask]> wrote:

Hi,

No, you should use the --tcustom option for this and provide a timing value for each slice (e.g., 0.5 0.25 0 -0.25 -0.5 0.5 0.25 0 -0.25 -0.5 0.5 0.25 0 -0.25 -0.5 etc)

All the best,

Mark

On 23 Dec 2017, at 08:03, huyang <[log in to unmask]> wrote:

Dear experts,

Slicetimer can be used for multi-band sequence data?

For example, the BOLD image has 15 slices,and three slices are scanned simutaneously. So I make a custom slice order file, which contains the exact scanning order of each slice ( 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5, one number per row) and I use slicetimer with "--ocustom" option (slicetimer -i input -o output --ocustom slice-order.txt). Is that right?

Thanks!

Yang Hu