| T O P I C R E V I E W |
| mathias.fleck |
Posted - Nov 02 2005 : 2:40:24 PM
The three subjects I have scanned since 10/27 have all shown a new problem: the first image on each functional run is all gray. This obviously makes for some realignment problems.
Daniel W. has also noticed this on all of his subjects since that date as well. |
| 12 L A T E S T R E P L I E S (Newest First) |
| mathias.fleck |
Posted - Nov 11 2005 : 11:18:31 AM
Thanks Jimmy!
Mine are all good now |
| jimmy.dias |
Posted - Nov 09 2005 : 11:46:40 PM
Hey all,
So we figured out what the problem is. One of the utilities that preps the 4.0T data by byteswapping the raw files had issues running on the Itanium. We've since fixed the problem, (lets say I'm 90% sure so I won't eat my words later on). I'm re-running everyone's data back through tonight and hopefully by the morning, you'll be looking at complete images throughout.
Cheers,
Jimmy |
| josh.bizzell |
Posted - Nov 09 2005 : 09:14:24 AM
For the data I saw, this problem occurred in the first slice of the first volume. There were very negative values (~ -32,000, mostly where the brain would be) and very positive values (~ 32,000) in the slice, which would cause the rest of the volume to look gray. All of the other slices in the volume looked normal (again, for the data I saw).
There are many temporary workarounds I could think of for this if the data needs to be analyzed (realigned, registered) quickly:
- Ignore the first volume completely
- Copy the first slice of the second volume to the first slice of the first volume
- Zero the first slice, if it's unimportant anatomically
|
| weissman |
Posted - Nov 09 2005 : 12:01:04 AM
My recent subjects from 11/8/2005 also have the same problem (e.g., 45387 and 45388).
Daniel |
| han |
Posted - Nov 08 2005 : 7:36:18 PM
My recent two subjects still have the same problem (subj ID 45380, 45386)
-Hans |
| ndennis |
Posted - Nov 08 2005 : 2:20:45 PM
same with my Ss 45376
-nancy |
| weissman |
Posted - Nov 08 2005 : 2:18:32 PM
Hi,
The problem that Matt Fleck mentioned on November 8th still hasn't been fixed.
As Matt said, the first slice of the first functional image from each run on the 4T is bad.
I just collected another subject today and the problem is still there.
Does anyone know when this problem will be fixed?
Thanks,
Daniel |
| carter |
Posted - Nov 08 2005 : 1:43:53 PM
Actually, same for 20051103_45377. |
| carter |
Posted - Nov 08 2005 : 1:29:52 PM
My first image on each functional for 20051108_45383 is strange too.
-Liz |
| mathias.fleck |
Posted - Nov 07 2005 : 11:23:23 AM
I'm still seeing same problem with my three subjects--
45360
45361
45366 |
| jimmy.dias |
Posted - Nov 06 2005 : 5:50:45 PM
Hey all,
Since the firewall had been installed, the machine running the recon had also been changed from Fourier to another machine with supposedly an updated version of the 4.0T recon. Unfortunately, this updated recon had the 'first image' problem that you all noticed.
I've reverted back to the older recon version for the 4.0T, the one that had been running on Fourier pre-firewall, and am now re-running all 4.0T data acquired since 10/27. All the data I have should be over in a couple of hours. Take a look and let me know if anything is off.
Cheers,
Jimmy |
| han |
Posted - Nov 03 2005 : 1:10:57 PM
I have the same issue from 10/27 |
