processing and Z cropping tiled Leica images

[daj1u06]

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

Hi all,

I have a user who is assessing bacterial biofilm clearance by ciliated airway
epithelial cells grown on air liquid interface cultures. He then fixes the cultures,
cuts out and divides the membrane and does FISH for the bacteria, immuno for
the cilia plus nuclear counterstain. The filter slices are mounted in mountant
between 2 coverslips and imaged on SP5 in tile scan mode, imaging a long, thin
radial strip of approx 45x1 fields at 512x512 resolution and multiple Z levels (up
to 50ish) It is impossible to get the filters perfectly flat - they tilt and buckle
so we need to set a large Z range to encompass absolute highest and lowest
points and many Z slices. Ultimately he wants to calculate an stimate of
biofilm volume

My 2 problems, seeking advice and solutions:
(1) because of the wide Z range, some slices on some fields cut into the filter
(autofluorescence and non specific binding at holes) or close to coverslip
(flash). Because of the undulations of the membrane, this varies from field of
view to field of view. Therefore I need a way to crop or blank Z slices off the
data set in a manner which is field of view specific, rather than global, but still
retains the relative Z position in the overall stack.

(2) Despite the fields forming an absolute linear strip, Leica's autostitcher
sometimes makes a complete hash of it. Is there any easy way to get a stitch
of Z stacks based simply on relative location.

We are talking reasonably large datafiles here (several GB)

Thanks in Advance,
Dave Johnston,
Biomedical Imaging Unit, Southampton, UK.

PS, learned the hard way, if doing tile sets, make sure scan head rotation is
set to 0 :-)

[G. Esteban Fernandez]

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

Have you tried ImageJ for stitching?  Fiji ImageJ (http://fiji.sc/)
comes pre-loaded with s stitching plugin where you can specify the
relative location of tiles to prime the autostitching algorithm
(Plugins > Stitching > Grid/Collection stitching).  It has worked very
well for me, the priming made a big difference and even stitched some
low-contrast brightfield images well.

-Esteban


On Thu, Aug 16, 2012 at 12:30 AM, David Johnston <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi all,
>
> I have a user who is assessing bacterial biofilm clearance by ciliated airway
> epithelial cells grown on air liquid interface cultures. He then fixes the cultures,
> cuts out and divides the membrane and does FISH for the bacteria, immuno for
> the cilia plus nuclear counterstain. The filter slices are mounted in mountant
> between 2 coverslips and imaged on SP5 in tile scan mode, imaging a long, thin
> radial strip of approx 45x1 fields at 512x512 resolution and multiple Z levels (up
> to 50ish) It is impossible to get the filters perfectly flat - they tilt and buckle
> so we need to set a large Z range to encompass absolute highest and lowest
> points and many Z slices. Ultimately he wants to calculate an stimate of
> biofilm volume
>
> My 2 problems, seeking advice and solutions:
> (1) because of the wide Z range, some slices on some fields cut into the filter
> (autofluorescence and non specific binding at holes) or close to coverslip
> (flash). Because of the undulations of the membrane, this varies from field of
> view to field of view. Therefore I need a way to crop or blank Z slices off the
> data set in a manner which is field of view specific, rather than global, but still
> retains the relative Z position in the overall stack.
>
> (2) Despite the fields forming an absolute linear strip, Leica's autostitcher
> sometimes makes a complete hash of it. Is there any easy way to get a stitch
> of Z stacks based simply on relative location.
>
> We are talking reasonably large datafiles here (several GB)
>
> Thanks in Advance,
> Dave Johnston,
> Biomedical Imaging Unit, Southampton, UK.
>
> PS, learned the hard way, if doing tile sets, make sure scan head rotation is
> set to 0 :-)

[George McNamara]

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

Hi Dave,

Contact your Leica confocal sales rep and have them visit to demo the
Leica MATRIX module(s) to do what you need. They might also want to
bring with them a Leica MATRIX application(s) expert. A quote for a new
64-bit PC might also be useful.

I've already complained to Leica that their LAS AF stage tiling
acquisition and image stitching is bad (I probably complained about it
being bad in TCS and LCS when I managed an SP1 - accomplished nothing).
Not that Zeiss LSM710 / ZEN software is any better.

George



On 8/16/2012 3:30 AM, David Johnston wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi all,
>
> I have a user who is assessing bacterial biofilm clearance by ciliated airway
> epithelial cells grown on air liquid interface cultures. He then fixes the cultures,
> cuts out and divides the membrane and does FISH for the bacteria, immuno for
> the cilia plus nuclear counterstain. The filter slices are mounted in mountant
> between 2 coverslips and imaged on SP5 in tile scan mode, imaging a long, thin
> radial strip of approx 45x1 fields at 512x512 resolution and multiple Z levels (up
> to 50ish) It is impossible to get the filters perfectly flat - they tilt and buckle
> so we need to set a large Z range to encompass absolute highest and lowest
> points and many Z slices. Ultimately he wants to calculate an stimate of
> biofilm volume
>
> My 2 problems, seeking advice and solutions:
> (1) because of the wide Z range, some slices on some fields cut into the filter
> (autofluorescence and non specific binding at holes) or close to coverslip
> (flash). Because of the undulations of the membrane, this varies from field of
> view to field of view. Therefore I need a way to crop or blank Z slices off the
> data set in a manner which is field of view specific, rather than global, but still
> retains the relative Z position in the overall stack.
>
> (2) Despite the fields forming an absolute linear strip, Leica's autostitcher
> sometimes makes a complete hash of it. Is there any easy way to get a stitch
> of Z stacks based simply on relative location.
>
> We are talking reasonably large datafiles here (several GB)
>
> Thanks in Advance,
> Dave Johnston,
> Biomedical Imaging Unit, Southampton, UK.
>
> PS, learned the hard way, if doing tile sets, make sure scan head rotation is
> set to 0 :-)
>
>    

[Barlow, Andrew]

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

Hi Dave,

We've recently implemented stitching of 3D, time resolved datasets in Volocity.
The relative positions of tiles is read from Leica lif files, which greatly speeds the process up (compared to blind stitching).
We've had some great feedback about this feature, please contact me offline if you'd like a demonstration.
Regards,
Andrew


JOIN US FOR OUR VOLOCITY AND COLUMBUS USER GROUP MEETINGS AT THE
EUROPEAN MICROSCOPY CONGRESS 2012
Manchester, UK, Friday 21st September 2012, 2pm - 6pm
Click here to register

Andrew Barlow PhD | Market Development Leader, CI & A
PerkinElmer | For the Better
[hidden email]
Phone:  + 44 2476 692229|   Fax:  [phone]   |   Mobile:   [phone]
Millburn Hill Road, Coventry, CV4 7HS, UK
www.perkinelmer.com
Please consider the environment before printing this e-mail.
This e-mail message and any attachments are confidential and proprietary to PerkinElmer, Inc.  If you are not the intended recipient of this message, please inform the sender by replying to this email or sending a message to the sender and destroy the message and any attachments.  Thank you.

> -----Original Message-----
> From: Confocal Microscopy List
> [mailto:[hidden email]] On Behalf Of David Johnston
> Sent: 16 August 2012 08:30
> To: [hidden email]
> Subject: processing and Z cropping tiled Leica images
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi all,
>
> I have a user who is assessing bacterial biofilm clearance by ciliated
> airway epithelial cells grown on air liquid interface cultures. He then
> fixes the cultures, cuts out and divides the membrane and does FISH for
> the bacteria, immuno for the cilia plus nuclear counterstain. The
> filter slices are mounted in mountant between 2 coverslips and imaged
> on SP5 in tile scan mode, imaging a long, thin radial strip of approx
> 45x1 fields at 512x512 resolution and multiple Z levels (up to 50ish)
> It is impossible to get the filters perfectly flat - they tilt and
> buckle so we need to set a large Z range to encompass absolute highest
> and lowest points and many Z slices. Ultimately he wants to calculate
> an stimate of biofilm volume
>
> My 2 problems, seeking advice and solutions:
> (1) because of the wide Z range, some slices on some fields cut into
> the filter (autofluorescence and non specific binding at holes) or
> close to coverslip (flash). Because of the undulations of the membrane,
> this varies from field of view to field of view. Therefore I need a way
> to crop or blank Z slices off the data set in a manner which is field
> of view specific, rather than global, but still retains the relative Z
> position in the overall stack.
>
> (2) Despite the fields forming an absolute linear strip, Leica's
> autostitcher sometimes makes a complete hash of it. Is there any easy
> way to get a stitch of Z stacks based simply on relative location.
>
> We are talking reasonably large datafiles here (several GB)
>
> Thanks in Advance,
> Dave Johnston,
> Biomedical Imaging Unit, Southampton, UK.
>
> PS, learned the hard way, if doing tile sets, make sure scan head
> rotation is set to 0 :-)

[Thomas Trusk]

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

I've been experimenting quite a bit with the latest incarnation of Leica's
stitching routine in LAS, and it is MUCH improved.  The automatic mode does
use content as a guide, so blank or sparsely populated images present a
problem.  If you turn AUTO off, the stitching is faster and edge-to-edge.

I should note that Leica offers a new option, that I have not had a chance to
see, that supposedly pre-maps focus over the entire area to be stitched.  I've
been told this function would allow one to scan only areas where regions of
interest exist, preventing the blank image problem and saving lots of storage
space.

Thomas C. Trusk, PhD
Director, Josh Spruill Imaging Facility
Regenerative Medicine and Cell Biology
Medical University of South Carolina
Charleston, SC  29425
http://mmi.musc.edu