DSI on a Philips 3T

69 views
Skip to first unread message

pgm...@gmail.com

unread,
Jan 17, 2017, 12:27:06 PM1/17/17
to DSI Studio, psu...@bangor.ac.uk, David Carey, e.kar...@bangor.ac.uk
Hi Frank, and others, 
first I would like to say thanks for sharing this software and your experience with DSI.  Secondly, we are looking to apply DSI on a philips 3T scanner, and would welcome any tips or pointers others may have.
As a first pass, we have used the dsi101_bmax4000_half.txt file available on the DSI studio site as a gradient direction file for our DTI scan (after reordering the columns - Philips wants the b value last).  We had a bit of a problem with recon memory at first, but if you turn on recon compression in the postprocessing tab, you can get around this.  We were then able to run, but as the default TR is for the longest B value - we found the time for the scan to be fairly long (40-46 mins for a scan that gives full coverage), so we reduced the number of scans to go a bit faster for this first run.  We should be able to analyse that scan next week and will let you know how it goes.
In an attempt to speed things up a bit (and prevent the recon memory problem) I then tried splitting the different B values for the DSI scan across three runs.  This did get rid of the recon memory problem, and also sped the scans up a bit (even for a full brain coverage) as the minimum TR is shorter for the scans with the lower b values. However, I'm wondering if this is appropriate, or if I should set the TE and TR for all the scans to be the same, and use the minimum TR from largest b value scan for all scans.  Would you suggest this as a way forward?  Or do you think DSI studio can cope with the DSI data split across three scans, all with different TEs and TRs?

Okay, I hope tht made sense.  Looking forward to learning more as we progress.

Paul.

Fang-Cheng Yeh

unread,
Jan 17, 2017, 8:10:06 PM1/17/17
to dsi-s...@googlegroups.com, psu...@bangor.ac.uk, David Carey, e.kar...@bangor.ac.uk
Hi Paul,

You can do either ways, but there are pros and cons. If TE and TR
are different, then you may need to acquire a b0 for each TE/TR
combination. The b0 images can be used to calculate the T1W and T2W
and eliminate all T1- and T2 shine through for all DWI. Using the
minimum TR of the largest b scan and applying it to other scans saves
this hassle, but the total scanning time will be longer. DSI Studio
once had this multiple TE/TR correction routine built-in but was
removed due to poor performance at low SNR area. I would recommend you
use the same TE/TR for all scans. As multi-band sequence is getting
more popular in recent years, the scanning time should be the last
factor to consider.

DSI data split can be readily handled in DSI Studio. I used to
acquire multiple sets of scans for a DSI on Bruker scanners and
combined all of them in DSI Studio. For Phillips scans, you may put
all DICOM files under the same folder, and DSI Studio should be able
to load all of them just like they were collected all together. Should
there be any question about merging the dataset, please feel free to
let me know.

Best regards,
Frank
> --
> You received this message because you are subscribed to the Google Groups
> "DSI Studio" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to dsi-studio+...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

Paul Mullins

unread,
Jan 19, 2017, 4:50:50 AM1/19/17
to dsi-s...@googlegroups.com
Thanks Frank, 

That all sounds good.  Using MB would be great, but unfortunately we do not yet have that capability on our system.  I may push Philips to see if we can get a WIP or something. and give it a try.  we might also try both methods of acquiring the data, and see how they compare.  Thanks once again, and I will let you know how we get on.

Cheers
Paul.


> For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "DSI Studio" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dsi-studio/7j2ed9Hd18Y/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dsi-studio+unsubscribe@googlegroups.com.

Fang-Cheng Yeh

unread,
Jan 19, 2017, 9:35:30 AM1/19/17
to dsi-s...@googlegroups.com
Glad to help here. Please feel free to let me know if there is
anything I can assist.

Best regards,
Frank
>> > email to dsi-studio+...@googlegroups.com.
>> > For more options, visit https://groups.google.com/d/optout.
>>
>> --
>> You received this message because you are subscribed to a topic in the
>> Google Groups "DSI Studio" group.
>> To unsubscribe from this topic, visit
>> https://groups.google.com/d/topic/dsi-studio/7j2ed9Hd18Y/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to
>> dsi-studio+...@googlegroups.com.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "DSI Studio" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to dsi-studio+...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages