How to Take More images per TR in Functional Scanning

Background
The limitation on the number of images that can be acquired per TR of scanning comes from two factors:
  1. The time required to create an image (about TE+20 msec for 128x64, and about TE+40 msec for 128x128)
  2. The amount of heat dissipated in the gradient coil

It turns out that by using fractional acquisition processing, we can achieve the same image resolution while acquiring only about half of the data. This results in about half the heat load on the gradient coils.

The steps below 'fool' the scanner into setting up fractional acquisitions. With it, I have successfully scanned up to 26 locations in a TR of 2500 msec.

What is the cost? As you reduce the fraction of actual k-space coverage, artifacts increase and SNR goes down.

On the Signa:

  1. Set up most of your functional scanning as usual, but set the TE to '8' on the Scan Timing page. To do this, you must enter the text in from the keyboard.
  2. When you set up the Scanning Range, you can safely anticipate a little more than one slice per 100 msec of TR. For a 2.5 second TR, for example, 26 slices works fine.
  3. As always, make sure that you select R/L Frequency Direction from the Acq Time page
  4. Once you have reached the Scan Ops page,type:
      Type:
        MCV (or press the [Modify CVs] on the plasma screen)

      Enter:

        IS_fnex [return]

      then

        0.75 [return] (you may go as low as 0.625 and have useable images...)

      Enter:

        opte [return]

      then

        42000 [return] (this is the desired TE in microseconds)

  5. Touch [Backup]

On the Instascan console:

No special setup required. You might want to switch to Research Mode (top left menu) and verify that the value of part_phase is being set to the value you entered, but this is optional

Scan as usual.

Back to HowTo pages
This page is maintained by MarkCohen [updated 08.09.00]