Open BenInglis opened 8 years ago
Ben, is there a way to check whether previously collected data used the correct settings? Could you check my sequences in my folder that contains the name "birdcage" (it's only one of them) and see if they were set correctly? I'm in the process of deciding which head coil to buy and trying to compare SNR between different options, so this information would be critical. Thanks!
If you always used a protocol with the birdcage already set then it remains set. I looked at your "birdcage" protocol and it was set correctly. We set this up after seeing the bug when we first tested the birdcage coil, fixing the problem for you but not for Derek or others who might have started with a different protocol.
To check the setting for any individual acquisition, the DICOM header includes an "unknown" field assigned as C:HCP for the birdcage (1H - circularly polarized coil). This will read C:HEA;HEP if either the 12ch or 32ch coil was used to receive, C:Body if the body coil was used to receive.
(Note that there is also a field, TransmitCoilName in the DICOM header. This is always assigned Body whether we are using 12ch, 32ch or birdcage coils to receive.)
An EPI scan that was set up to use the 12ch or 32ch coil and saved into a protocol with HEA, HEP set under the System tab is likely to default to using the Body coil as the receiver rather than the birdcage coil. This will produce images with poor SNR but that otherwise resemble birdcage coil EPIs.
The solution is to always build protocols from EPI scans that have previously been acquired with the birdcage coil and have that coil selected already under the System tab. Also, it's a good idea to get into the habit of checking the System tab for the RF receive coil selection before using an EPI protocol for the first time with the birdcage coil.
This default body coil selection behavior is a bug that comes from having a custom coil file for the (custom) birdcage. It is in the same family as the "protocol has been fixed" error that arises when a protocol established for the 32ch coil is set up for use with the 12ch coil, or vice versa. In this latter case the bug also causes GRAPPA, if enabled, to turn off.