Uneven number of control-label frames

[Lina Jarutytė]

Hi,

If there's an uneven number of control-label frames, how one should check if it's due to incomplete pairs or M0 image in time-series?

For example, I have a case with 2D EPI PASL images (no background suppression) acquired with Siemens scanner and, according to the protocol, there should be no M0 image. 

Any advise how to deal with this issue would be highly appreciated.

Many thanks!

Lina

[jan.honza.petr]

Hi Lina,

That is pretty difficult as most vendors do not save this information to the DICOM. This really depends if you have Background suppression. If yes, then this should be turned of for M0, making it easily recognizable.

Sometimes, without BS, the M0s have a different TR making it also distinct to other control/labels.

You can also try the following to extract it from DICOM header:

Philips (not Siemens though) sometimes saves LABEL or CONTROL information to the private tag 'Private_2005_1429'

Siemens saves a lot of information to this large string: 'Private_0029_1020'.

I do not remember though, if the CONTROL/LABEL information is there and most of my Siemens DICOMs have this whole structure removed for anonymization as there is a mix of private and technical information.

You would have to send me the dump from DICOM stripped of patient data.

In your case (old Siemens product sequence with 2D EPI PASL without BS and with Q2TIPS), it was typical to acquire odd number of control labels - CLCLC or LCLCL. Typically, the first one is thrown away.

regards,

Jan