|
diaz
BIAC Alum
USA
212 Posts |
 Posted - Jan 27 2006 : 3:39:49 PM
|
Dear BIAC Users:
The 3T has a new and different design in which a vacuum is established to remove the air surrounding the gradient coils. This vacuum design reduces the noise generated by the gradients, and ensures that the 3T meets the new FDA noise limits.
An unfortunate side effect has been that the new gradients do not dissipate heat as well as the gradients on our 4T system. To protect the gradients, a temperature sensor will shut down the scanner if the system is overheating. This is not an issue for routine scanning; but some of our research spiral protocols with large numbers of slices and long runs have caused the temperature sensor to shutdown the acquisition. As a consequence, General Electric has required us to reduce the throughput on spiral scans performed on the 3T scanner.
We are currently requiring users to collect no more than 20 slices per second when using spiral acquisitions. If we still have overheating problems, then we may further reduce the acquisition rate to 18 slices per second. Echoplanar sequences are not affected by this rule, because they put less stress on the gradients and because they can only acquire data at a maximum of 17 slices per second.
If you have been acquiring data at greater than 20 slices per second, you will have to modify your acquisition sequence. There are different ways to compensate, and the choice depends upon your experimental goals. One way is to reduce your TR – for example, using a TR of 2 sec allows you to collect up to 40 slices per volume. Since the whole brain can be acquired in 34-36 slices with a slice thickness of 3.75 mm, it is possible to acquire isotropic (cubic) voxels of 3.75 mm with a 2 sec TR. If your experimental question requires a shorter TR than 2 sec, then you either reduce the number of slices and sacrifice coverage, or you can increase slice thickness and maintain coverage.
Please make these changes now. The MR Technologists have been instructed not to let anyone exceed these limits – we must protect the equipment for all investigators.
Finally, we note that it is unusual within the fMRI field to collect data with the very high throughputs that have become typical at the BIAC. Indeed, in many cases the high throughput acquisition rates used here have been unnecessary to answer the research questions posed. Anyone needing advice on how to adjust their protocols can contact Allen Song or Gregory McCarthy.
|
Michele T. Diaz, Ph.D.
Associate Director
Brain Imaging and Analysis Center |
Edited by - gregory.mccarthy on Jan 27 2006 6:01:30 PM |
|
Scanning at 3T
Forum Locked
