Dear Bruce,
Best wishes for 2014!
I will heed your advice and cc to the list. Indeed, the MGH protocol I obtained from Andre has one acquisition (NSA). Related, I cc this post:
We have been trying adaptations of the T2spaceFLAIR MGH protocol on our Phillips 3T scanners (sequences called VISTA). Incorporation of the T2 space FLAIR has been mostly successful at editing out the dura from the pial surface. The only concern are minute differences occasionally appearing as small indentations (at most times <1-2mm) between the woFLAIR.pial and the final pial surfaces, including in intrasulcal (duraless) areas; the amount of these small mismatches is variable from patient to patients, with some trials with almost perfect overlap, and others with visible, if small, differences. I wonder if this has been your experience too, and if this is due mostly to bb registration limits or to differences in pixel bandwidth between mprage and space flair sequences (in our case, see below). In FSwiki, the recommendation is that T2 space flair “should be bandwidth, geometry and readout matched to the mprage”.
1. Speaking about pixel bandwidth, logistics most likely would not allow application of a MEMPR sequence on our Phillips. According to Siemens MPRAGE MGH specs, "exceptions indicate that if a multiple echo sequence is not available, one should choose a bandwidth of 195 Hz/pixel for the MPRAGE". In that case/ using the Siemens Trio MPRAGE protocol as if the MEMPR is not available, how one would adjust the bandwidth in T2spaceFLAIR protocol (currently ~650 in the Siemens space flair and MEMPR protocols) to "match" it to the "simple" MPRAGE protocol (decreasing it to 195 ?). Would it be a workaround? (Please comment on the Siemens case, obviously, I do not expect any light on Phillips without further testing). 2. Could you clarify what geometry matched means? Please advise,
Thank you,
Octavian
On Fri, Jan 3, 2014 at 9:12 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Octavian
can you cc the list so that others can answer? What is NSA? We don't have a ton of experience with it, but 1 has seemed ok (not sure if we've ever tried two) Bruce
On Thu, 2 Jan 2014, Octavian Lie wrote:
Dear Bruce,
Thank you. We used one vs 2 NSA T2 space FLAIR sequences, without a consistent difference in results, with the 2-NSA run obviously doubling the scan time. Is there a preferred number of NSA you recommend, or we should be fine with one? Both are at resolution 1X1X1 mm. Thank you,
Octavian
On Thu, Jan 2, 2014 at 11:56 AM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote: Hi Octaviaan no, not really. You should motion correct and average them using something like mri_motion_correct*
We do typically bandwidth/readout match them to the memprages so that there is no differential distortion cheers Bruce On Wed, 1 Jan 2014, Octavian Lie wrote: Dear All, We recently implemented pial correction using T2 space FLAIR with freesurfer v 5.3, mostly successfully. We use 2-3 MPRAGE (sagittal + axial and/or coronal) volumes as 001.mgz, 002/003.mgz for the motion correction step in recon-all pipeline. The typical times for acquiring space FLAIR in our settings vary from 2.5 to 6 min, depending on the resolution/NSA no. I was wondering how freesurfer deals with motion artifacts in the space FLAIR sequence (especially with longer acquisition times), and if there is a strategy to minimize motion distortion with these (such as getting second runs for averaging, etc). Please advise, Octavian.The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Hi Octavian
it's not so easy to reduce the bandwidth in the T2-space sequences, but certainly you can try. Geometry-matched just means that the FOV and matrix sizes are the same.
cheers Bruce On Fri, 3 Jan 2014, Octavian Lie wrote:
Dear Bruce,
Best wishes for 2014!
I will heed your advice and cc to the list. Indeed, the MGH protocol I obtained from Andre has one acquisition (NSA). Related, I cc this post:
We have been trying adaptations of the T2spaceFLAIR MGH protocol on our Phillips 3T scanners (sequences called VISTA). Incorporation of the T2 space FLAIR has been mostly successful at editing out the dura from the pial surface. The only concern are minute differences occasionally appearing as small indentations (at most times <1-2mm) between the woFLAIR.pial and the final pial surfaces, including in intrasulcal (duraless) areas; the amount of these small mismatches is variable from patient to patients, with some trials with almost perfect overlap, and others with visible, if small, differences. I wonder if this has been your experience too, and if this is due mostly to bb registration limits or to differences in pixel bandwidth between mprage and space flair sequences (in our case, see below). In FSwiki, the recommendation is that T2 space flair “should be bandwidth, geometry and readout matched to the mprage”.
- Speaking about pixel bandwidth, logistics most likely would not
allow application of a MEMPR sequence on our Phillips. According to Siemens MPRAGE MGH specs, "exceptions indicate that if a multiple echo sequence is not available, one should choose a bandwidth of 195 Hz/pixel for the MPRAGE". In that case/ using the Siemens Trio MPRAGE protocol as if the MEMPR is not available, how one would adjust the bandwidth in T2spaceFLAIR protocol (currently ~650 in the Siemens space flair and MEMPR protocols) to "match" it to the "simple" MPRAGE protocol (decreasing it to 195 ?). Would it be a workaround? (Please comment on the Siemens case, obviously, I do not expect any light on Phillips without further testing). 2. Could you clarify what geometry matched means? Please advise,
Thank you,
Octavian
On Fri, Jan 3, 2014 at 9:12 PM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote:
Hi Octavian
can you cc the list so that others can answer? What is NSA? We don't have a ton of experience with it, but 1 has seemed ok (not sure if we've ever tried two) Bruce
On Thu, 2 Jan 2014, Octavian Lie wrote:
Dear Bruce,
Thank you. We used one vs 2 NSA T2 space FLAIR sequences, without a consistent difference in results, with the 2-NSA run obviously doubling the scan time. Is there a preferred number of NSA you recommend, or we should be fine with one? Both are at resolution 1X1X1 mm. Thank you,
Octavian
On Thu, Jan 2, 2014 at 11:56 AM, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote: Hi Octaviaan no, not really. You should motion correct and average them using something like mri_motion_correct*
We do typically bandwidth/readout match them to the memprages so that there is no differential distortion cheers Bruce On Wed, 1 Jan 2014, Octavian Lie wrote: Dear All, We recently implemented pial correction using T2 space FLAIR with freesurfer v 5.3, mostly successfully. We use 2-3 MPRAGE (sagittal + axial and/or coronal) volumes as 001.mgz, 002/003.mgz for the motion correction step in recon-all pipeline. The typical times for acquiring space FLAIR in our settings vary from 2.5 to 6 min, depending on the resolution/NSA no. I was wondering how freesurfer deals with motion artifacts in the space FLAIR sequence (especially with longer acquisition times), and if there is a strategy to minimize motion distortion with these (such as getting second runs for averaging, etc). Please advise, Octavian.The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
Octavian Lie