Hi,
I would really not recommend using PNM without triggers, based on the experience in our lab, as relative drifts in timing between the physiological data and the MR scans is quite common. Even though the drifts are often small, over the length of the scan it is often enough to be problematic (as any mismatch in timing that is a substantial fraction of the cardiac cycle will stop the regression working well).
If you have recorded scanner triggers then PNM can get very accurate timing. Whether you need to adjust for deleted volumes or not will depend on whether your scanner triggers only occur after the initial dummy volumes or if they are also present for these. This will depend on the scanner manufacturer and possibly other software/hardware configurations. It is quite common, I believe, that the triggers are only present for the volumes after the initial dummy ones (the scans used to remove the saturation effects at the beginning of the run). In your example, if you did not have any triggers until volume 6, then nothing would need to be done. Otherwise you would need to remove the triggers before your 6th scan.
I hope this is helpful.
All the best,
Mark
On 7 Nov 2012, at 12:23, SUBSCRIBE FSL zzang <[log in to unmask]> wrote:
> Hi Mark,
>
> thanks for the reply.
> actually we couldn't record the scanner triggers, we can only calculate the time based on the TR.
> are you suggesting that PMN code can align the physiological data with the functional data appropriately even they start and end at different time point, as long as the timing for triggers is accurate?
>
> for example, we will remove the first 5 volumes of our functional data and the corresponding trigger timing would then starts from
> the time of the 6th TR. should we set the 6th TR as zero, thus the beginning, or just leave it as the original time?
>
> Best,
>
> Wei
>
|