*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Craig,

What exactly is image registration?  

At one point we thought of just oversampling in Z and then adjusting the
image digitally post-acquisition, depending on the Z-shifts.  This is
impractical for our application because we are looking at very small,
low-signal vesicles moving through the cell.  So oversampling would cause
issues with bleaching and we would lose too much time resolution.

If image registration is something else, then can you explain?

Thanks,

Dan

On Mon, 31 Jan 2011 10:23:06 -0700, Craig Brideau <[log in to unmask]>
wrote:

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>From a software side you can also just image volumes and then use image
>registration to undo the drifts in the data.  This may not be optimal for
>your application though.
>
>Craig
>
>
>On Mon, Jan 31, 2011 at 10:22 AM, Craig Brideau <[log in to unmask]>wrote:
>
>> Box in your scope; that is, build an enclosure around it.  The box will
>> block any air currents or other disturbances.  If temperature drift in the
>> room is more than 1 degree C you can also add temperature control to the box
>> to thermally isolate the microscope from the room.  Finally, what sort of
>> table is the microscope sitting on?
>>
>> Craig
>>
>>
>> On Mon, Jan 31, 2011 at 9:49 AM, Daniel Murphy <
>> [log in to unmask]> wrote:
>>
>>> *****
>>> To join, leave or search the confocal microscopy listserv, go to:
>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>> *****
>>>
>>> Hello,
>>>
>>> Some advice and a plea for help on XYZ drifts!
>>>
>>> We have been doing time-lapse studies on our Zeiss LSM 5 Live, Axiovert
>>> 200M
>>> inverted scope.  Typically these go for 20min durations at 1frame/30sec.
>>>  We
>>> capture z-stacks that are typically 10-15 slices thick on a 40X Water with
>>> 2.5X optical zoom.  We also use a Warner instruments micro-incubation
>>> system
>>> DH40i (http://www.warneronline.com/product_info.cfm?id=1165) with heated
>>> dish cover to maintain a temp of 37C.
>>>
>>> We have had persistent issues with drifts in X, Y and Z.  Initially the
>>> problems were sever, with shifts of up to a few microns in all 3
>>> directions.
>>>  We have made progressive adjustments to our protocol with some big
>>> improvements, but the problem is still significant.
>>>
>>> First, we found that by surrounding the open area below the stage (where
>>> you
>>> can access the objectives) with seran wrap, this provided a good way of
>>> protecting the system from air currents and thermal influence from the
>>> environment.  We also characterized the incubation system and found it
>>> works
>>> better to run it without feedback at a constant voltage (the feedback
>>> response was too slow because the incubation system is too much of a heat
>>> sink).  We typically turn on the incubator and let it equilibrate for at
>>> least 15min (with the dish inside as well, whenever possible).
>>>
>>> XYZ drift still remains, however.  It seems to come and go.  For one
>>> experiment it will be almost unnoticeable, but for another it will make
>>> the
>>> data practically unusable.  Sometimes the drifts are just in one
>>> direction.
>>>  Other times the stack shifts in Z up and down several times in one time
>>> sequence.  It seems to be very irregular and so probably due to random
>>> fluctuations in the environment.
>>>
>>> XY shifts are not too bad, so long as the area of interest remains in
>>> view.
>>>  There are several ways to adjust for the shift post-acquisition --- you
>>> have more wiggle room since there are all those pixels in every direction.
>>> Z shifts are the real issue that plagues us.  Any advice or ideas would be
>>> warmly welcomed.
>>>
>>> Daniel Murphy
>>> Albert Einstein College of Medicine
>>> Optical Imaging Manager, Cell and Molecular Neuroimaging Core
>>> 1410 Pelham Parkway South
>>> Bronx, NY 10461
>>> Ph 718-430-4027/8985
>>>
>>
>>