Re: Objective lens chromatic aberration - Color shift correction for colocalization analysis - was Glycerol Objectives - experience with

[Stanislav Vitha]

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

Hi Dan,
The translational correction is very useful,  but what if the shift varies
across the field?
Do you know if there is a plugin or a script to perform affine transform in
3D? The goal would be to "warp" one channel in XYZ to match the second channel.
I have imaged 100 nm multi-color beads on our confocal (FV1000 with a
hand-picked 100x/1.4 Plan Super Apo lens), and besides a moderate z-shift
between blue and other channels (as expected), I have a lateral chromatic
aberration between the "DAPI" channel (405 nm ex, 430-460 em) and the longer
wavelength channels (green, orange, red fluorescence). So for instance, the
colors are colocalized in top left corner of the image, but are several
pixels off in XY plane in bottom right of the image. The UnwarpJ plugin for
ImageJ works well to correct this in individual XY images, but 3D correction
may be needed, since the plan correction for the blue channel (405 nm laser
illumination) is not as good as for the rest of the spectrum.  

 

Stan Vitha
Microscopy and Imaging Center
Texas A&M University


On Tue, 11 Jan 2011 11:23:02 +0100, Daniel James White <[hidden email]> wrote:

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>Hi Martin,
>
>On Jan 11, 2011, at 7:01 AM, CONFOCALMICROSCOPY automatic digest system wrote:
>
>> Date:    Mon, 10 Jan 2011 17:18:29 -0600
>> From:    Martin Wessendorf <[hidden email]>
>> Subject: Re: Glycerol Objectives - experience with
>>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>>
>> On 1/10/2011 4:32 PM, Rosemary White wrote:
>>
>>> I've found that the red and blue emission aren't quite lined up vertically,
>>> at least with our "blue" 63x objective, so have to cut the top one or two
>>> slices off a blue vertical series (depending on depth of slice), and the
>>> bottom slices off a red series and reassemble to get the emissions aligned -
>>> i.e. from the same depth in the tissue.  The objectives do vary a bit,
>>> perhaps ours isn't as well corrected as some.
>>
>> Can anyone explain where the difficulty arises in correcting for axial
>> chromatic aberration?  I've consistently seen serious axial chromatic
>> aberration in good quality oil objectives from one very reputable
>> manufacturer, between red, green and far-red (1 um off in the z-axis
>> between green and red, and between red and far-red; 2 um between green
>> and far-red).  I have always heard that axial chromatic aberration was
>> easy to correct for and would've thought that a solution for 3-color
>> correction would've been found 100 years ago.  However, I don't know
>> enough optics to understand the subtleties.  Are there trade-offs to
>> trying to obtain good correction, besides the cost in scattering of
>> adding additional lens elements?
>
>Sadly, not all objective lenses were made equal.
>
>In our hands on point scanning confocals, even when you align the

pinhole(s) as best you can
>there can still be a micron of misalignment in between "green", "red" or
"far red",
>with DAPI etc. often being way off, as it comes through a different optic
fiber and collimator anyway (eg on an LSM 510)

>
>I have tested different lenses of the same spec from the same manufacturer,
>and found that they all have their own personality.
>Good correction is very possible, but "good" is a relative thing.
>
>I send back ones that are too "bad", and keep the "good" ones.
>I take z stacks of 1 micron multi colour  beads to measure the remaining error
>(dont need sub resolution bead images for this measurement)
>
>There will always be a significant error, even in the very best
>"Super-dooper Mega Extra Apo-Chromat"  lens you can get from any

manufacturer for less than 100 k dollars.
>They simply can not be made perfect at a reasonable cost (as explained to
me by a Zeiss lens guru)
>
>If you want to do precise colocalization studies at the highest optical
resolution (everyone does, even if they don't immediately realize it)
>then you simply MUST measure the error, then correct/shift the images (in
3D) before colocalization/correlation analysis.
>
>This can be done in 3D in ImageJ/ Fiji
>using nice interpolation methods from Erik Meijering, for sub pixel shifts
- TJ-Translate :
>http://pacific.mpi-cbg.de/wiki/index.php/TransformJ
>leading to
>http://imagescience.org/meijering/software/transformj/translate.html
>
>or in the great colour shift corrector of Huygens Professional (no
commercial interest - just a happy customer)
>http://www.svi.nl/ChromaticShiftCorrector
>
>or roughly in the Zeiss AIM 510 software in xy only (whole pixel shifts),
>
>The effect on the 2 channel scatterplot / 2D histogram / fluorogram is very
significant.
>A ugly poorly correlated cloud turns into tight correlated populations of
pixels,

>and the coloc. coefficients jump much higher.
>
>more info here:
>http://ifn.mpi-cbg.de/wiki/ifn/index.php/Chromatic_aberration_measurement_and_correction
>
>cheers
>
>Dan
>
>
>
>>
>> Thanks--
>>
>> Martin
>> --
>> Martin Wessendorf, Ph.D.                   office: [phone]
>> Assoc Prof, Dept Neuroscience                 lab: [phone]
>> University of Minnesota             Preferred FAX: [phone]
>> 6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
>> Minneapolis, MN  55455                    e-mail: [hidden email]
>
>Dr. Daniel James White BSc. (Hons.) PhD
>Senior Microscopist / Image Visualisation, Processing and Analysis
>Light Microscopy and Image Processing Facilities
>Max Planck Institute of Molecular Cell Biology and Genetics
>Pfotenhauerstrasse 108
>01307 DRESDEN
>Germany
>
>[phone] (German Mobile)
>[phone] (Work phone at MPI-CBG)
>[phone] (Fax MPI-CBG LMF)
>
>http://www.bioimagexd.net  BioImageXD
>http://pacific.mpi-cbg.de                Fiji -  is just ImageJ (Batteries Included)
>http://www.chalkie.org.uk                Dan's Homepages
>https://ifn.mpi-cbg.de  Dresden Imaging Facility Network
>dan (at) chalkie.org.uk
>( white (at) mpi-cbg.de )

[Daniel James White]

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

Hi Stan,

On Jan 14, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system wrote:

>
> Date:    Thu, 13 Jan 2011 15:28:40 -0600
> From:    Stanislav Vitha <[hidden email]>
> Subject: Re: Objective lens chromatic aberration - Color shift correction for colocalization analysis - was Glycerol Objectives - experience with
>
> Hi Dan,
> The translational correction is very useful,  but what if the shift varie=
> s
> across the field?

then the assumption of the shift being space invariant is clearly false,
and you need a more clever way to correct it, that IS spatially variant... read on.

> Do you know if there is a plugin or a script to perform affine transform =
> in
> 3D?

Oh yes!

> The goal would be to "warp" one channel in XYZ to match the second ch=
> annel.

There are options here...

In Fiji (i just ImageJ) there is Ignacio's nice bUnwarpJ
http://pacific.mpi-cbg.de/wiki/index.php/BUnwarpJ
points to
http://biocomp.cnb.csic.es/~iarganda/bUnwarpJ/index.html

dont think it can also warp z though...

> I have imaged 100 nm multi-color beads on our confocal (FV1000 with a
> hand-picked 100x/1.4 Plan Super Apo lens), and besides a moderate z-shift=
>
> between blue and other channels (as expected), I have a lateral chromatic=
>
> aberration between the "DAPI" channel (405 nm ex, 430-460 em) and the lon=
> ger
> wavelength channels (green, orange, red fluorescence). So for instance, t=
> he
> colors are colocalized in top left corner of the image, but are several
> pixels off in XY plane in bottom right of the image. The UnwarpJ plugin f=
> or
> ImageJ works well to correct this in individual XY images,

not the same as bUnwarpJ, but similar....

> but 3D correct=
> ion
> may be needed, since the plan correction for the blue channel (405 nm las=
> er
> illumination) is not as good as for the rest of the spectrum.=20=20=20

thats often the case,
as well as the UC laser coming in through a different collimator, so its often 0.5 microns or so off the vis lines.

cheers


Dan


>
> =20
>
> Stan Vitha
> Microscopy and Imaging Center
> Texas A&M University

Dr. Daniel James White BSc. (Hons.) PhD
Senior Microscopist / Image Visualisation, Processing and Analysis
Light Microscopy and Image Processing Facilities
Max Planck Institute of Molecular Cell Biology and Genetics
Pfotenhauerstrasse 108
01307 DRESDEN
Germany

[phone] (German Mobile)
[phone] (Work phone at MPI-CBG)
[phone] (Fax MPI-CBG LMF)

http://www.bioimagexd.net  BioImageXD
http://pacific.mpi-cbg.de                Fiji -  is just ImageJ (Batteries Included)
http://www.chalkie.org.uk                Dan's Homepages
https://ifn.mpi-cbg.de  Dresden Imaging Facility Network
dan (at) chalkie.org.uk
( white (at) mpi-cbg.de )

[James Pawley]

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

Hi all,

Chromatic aberration is a measure of how the focal length (f) of a
lens changes with wavelength. As the magnification of any optical
system depends on the focal length of the lenses involved, of course
this means that, to the extent that chromatic aberration is present,
the magnification will change with wavelength. In simple optical
systems it is visible as the pronounced red or blue fringes around
the edges of images of high-contrast features. This is called lateral
chromatic aberration and, if the optical system is properly aligned,
it "should" be essentially invisible in the center of the field and
get larger as you approach the edges. (The finding that it lines up
in one corner but not elsewhere makes me think that there may be an
alignment issue.)

As mentioned before all refractive lenses have some chromatic
aberration. (Reflective lenses do not). A good plan-apo, may be
corrected for 4 wavelengths but the wavelengths chosen may vary from
lens to lens. Between these wavelengths there will show
wavelength-dependent shifts in focal length. The manufacturer
endeavors to keep these shifts less than the depth of field of the
objective so that they are not annoying to the naked eye when viewing
multicolored specimens. In practice you will only see the effects
when viewing planar objects such as patterns in thin metal films that
have been evaporated onto the glass surface.

Alan Boyde was the first to point out that these variations in f were
found in even the most expensive objectives, using a tandem-disk
scanner with Hg illumination (which has a number of intensity peaks
in the visible) and making backscattered/reflected light images of
planar features (surface of the slide). In fact he used this
"problem" to make the first truly live-time 3D confocal microscope:
Lenses were designed to have pronounced linear chromatic aberration.
Consequently, different planes in the specimen satisfied the confocal
conditions at different wavelengths. He then used prisms or
diffraction elements in the oculars that were oriented so that, say,
the red image was slightly displaced to the inside (towards the other
eye) of the blue. Looking through the oculars, the viewer then saw a
stereo image, obtained at full NA that was also color-coded for
topographic height. Because confocal images could be obtained at
several focal depths simultaneously, it was no longer necessary to
move the specimen in z to obtain a focal series, making live-stereo
images possible. These livetime stereo images could even be recorded
on tape and shown to groups using special glasses

 From this you can see that any attempt to correct for CA must be able
to respond to changes in magnification with wavelength as well as the
vertical displacements between the z-location of the on-axis peak
confocal signal vs wavelength. As noted above, there may also be
misalignment issues.

Anyone still willing to try to undertake this task should spend a lot
of time studying the work of Cremer and Cremer in the 1980s and
1990s. These scientists demonstrated one of the first effective
systems of super-resolution widefield and confocal imaging as applied
to FISH (fluorescent in-sit hybridization). They labelled one gene on
a chromosome red, and another nearby one green. They then recorded
the green spot and found its centroid. DItto, the red spot. Because
these two spots were separated by wavelength, their images did not
overlap in the recording system and therefore the distance between
them could be measured with an accuracy that depended on the ability
to determine the position of the centroid of each spot, rather than
on the Abbe resolution of the optics. However, as the position of the
centroid did depend on the CA of the optical system, they had to make
very careful measurements of this parameter in order to apply the
proper correction factors.

Hope to see some of you in Vancouver next June!

Cheers,

Jim Pawley

***************************************************************************
Prof. James B. Pawley,                          Ph.
608-238-3953                        
21. N. Prospect Ave. Madison, WI 53726 USA
[hidden email]
3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/            Applications due by March 15, 2011
               "If it ain't diffraction, it must be statistics." Anon.

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>Hi Dan,
>The translational correction is very useful,  but what if the shift varies
>across the field?
>Do you know if there is a plugin or a script to perform affine transform in
>3D? The goal would be to "warp" one channel in XYZ to match the
>second channel.
>I have imaged 100 nm multi-color beads on our confocal (FV1000 with a
>hand-picked 100x/1.4 Plan Super Apo lens), and besides a moderate z-shift
>between blue and other channels (as expected), I have a lateral chromatic
>aberration between the "DAPI" channel (405 nm ex, 430-460 em) and the longer
>wavelength channels (green, orange, red fluorescence). So for instance, the
>colors are colocalized in top left corner of the image, but are several
>pixels off in XY plane in bottom right of the image. The UnwarpJ plugin for
>ImageJ works well to correct this in individual XY images, but 3D correction
>may be needed, since the plan correction for the blue channel (405 nm laser
>illumination) is not as good as for the rest of the spectrum.  
>
>
>
>Stan Vitha
>Microscopy and Imaging Center
>Texas A&M University
>
>
>On Tue, 11 Jan 2011 11:23:02 +0100, Daniel James White
><[hidden email]> wrote:
>
>>*****
>>To join, leave or search the confocal microscopy listserv, go to:
>>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>*****
>>
>>Hi Martin,
>>
>>On Jan 11, 2011, at 7:01 AM, CONFOCALMICROSCOPY automatic digest
>>system wrote:
>>
>>>  Date:    Mon, 10 Jan 2011 17:18:29 -0600
>>>  From:    Martin Wessendorf <[hidden email]>
>>>  Subject: Re: Glycerol Objectives - experience with
>>>
>>>  *****
>>>  To join, leave or search the confocal microscopy listserv, go to:
>>>  http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>>  *****
>>>
>>>  On 1/10/2011 4:32 PM, Rosemary White wrote:
>>>
>>>>  I've found that the red and blue emission aren't quite lined up
>>>>vertically,
>>>>  at least with our "blue" 63x objective, so have to cut the top one or two
>>>>  slices off a blue vertical series (depending on depth of slice), and the
>>>>  bottom slices off a red series and reassemble to get the
>>>>emissions aligned -
>>>>  i.e. from the same depth in the tissue.  The objectives do vary a bit,
>>>>  perhaps ours isn't as well corrected as some.
>>>
>>>  Can anyone explain where the difficulty arises in correcting for axial
>>>  chromatic aberration?  I've consistently seen serious axial chromatic
>>>  aberration in good quality oil objectives from one very reputable
>>>  manufacturer, between red, green and far-red (1 um off in the z-axis
>>>  between green and red, and between red and far-red; 2 um between green
>>>  and far-red).  I have always heard that axial chromatic aberration was
>>>  easy to correct for and would've thought that a solution for 3-color
>>>  correction would've been found 100 years ago.  However, I don't know
>>>  enough optics to understand the subtleties.  Are there trade-offs to
>>>  trying to obtain good correction, besides the cost in scattering of
>>>  adding additional lens elements?
>>
>>Sadly, not all objective lenses were made equal.
>>
>>In our hands on point scanning confocals, even when you align the
>pinhole(s) as best you can
>>there can still be a micron of misalignment in between "green", "red" or
>"far red",
>>with DAPI etc. often being way off, as it comes through a different optic
>fiber and collimator anyway (eg on an LSM 510)
>>
>>I have tested different lenses of the same spec from the same manufacturer,
>>and found that they all have their own personality.
>>Good correction is very possible, but "good" is a relative thing.
>  >
>>I send back ones that are too "bad", and keep the "good" ones.
>>I take z stacks of 1 micron multi colour  beads to measure the
>>remaining error
>>(dont need sub resolution bead images for this measurement)
>>
>>There will always be a significant error, even in the very best
>>"Super-dooper Mega Extra Apo-Chromat"  lens you can get from any
>manufacturer for less than 100 k dollars.
>>They simply can not be made perfect at a reasonable cost (as explained to
>me by a Zeiss lens guru)
>>
>>If you want to do precise colocalization studies at the highest optical
>resolution (everyone does, even if they don't immediately realize it)
>>then you simply MUST measure the error, then correct/shift the images (in
>3D) before colocalization/correlation analysis.
>>
>>This can be done in 3D in ImageJ/ Fiji
>>using nice interpolation methods from Erik Meijering, for sub pixel shifts
>- TJ-Translate :
>>http://pacific.mpi-cbg.de/wiki/index.php/TransformJ
>>leading to
>>http://imagescience.org/meijering/software/transformj/translate.html
>>
>>or in the great colour shift corrector of Huygens Professional (no
>commercial interest - just a happy customer)
>>http://www.svi.nl/ChromaticShiftCorrector
>>
>>or roughly in the Zeiss AIM 510 software in xy only (whole pixel shifts),
>>
>>The effect on the 2 channel scatterplot / 2D histogram / fluorogram is very
>significant.
>>A ugly poorly correlated cloud turns into tight correlated populations of
>pixels,
>>and the coloc. coefficients jump much higher.
>>
>>more info here:
>>http://ifn.mpi-cbg.de/wiki/ifn/index.php/Chromatic_aberration_measurement_and_correction
>>
>>cheers
>>
>>Dan
>>
>>
>>
>>>
>>>  Thanks--
>>>
>>>  Martin
>>>  --
>>>  Martin Wessendorf, Ph.D.                   office: [phone]
>>>  Assoc Prof, Dept Neuroscience                 lab: [phone]
>>>  University of Minnesota             Preferred FAX: [phone]
>>>  6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
>>>  Minneapolis, MN  55455                    e-mail: [hidden email]
>>
>>Dr. Daniel James White BSc. (Hons.) PhD
>>Senior Microscopist / Image Visualisation, Processing and Analysis
>>Light Microscopy and Image Processing Facilities
>>Max Planck Institute of Molecular Cell Biology and Genetics
>>Pfotenhauerstrasse 108
>>01307 DRESDEN
>>Germany
>>
>>[phone] (German Mobile)
>>[phone] (Work phone at MPI-CBG)
>>[phone] (Fax MPI-CBG LMF)
>>
>>http://www.bioimagexd.net        BioImageXD
>>http://pacific.mpi-cbg.de                Fiji -  is just ImageJ
>>(Batteries Included)
>>http://www.chalkie.org.uk                Dan's Homepages
>>https://ifn.mpi-cbg.de                        Dresden Imaging
>>Facility Network
>>dan (at) chalkie.org.uk
>>( white (at) mpi-cbg.de )

--
***************************************************************************
Prof. James B. Pawley,                          Ph.
608-238-3953                        
21. N. Prospect Ave. Madison, WI 53726 USA
[hidden email]
3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/            Applications due by March 15, 2011
               "If it ain't diffraction, it must be statistics." Anon.

[Andreas Bruckbauer]

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

 

> I have imaged 100 nm multi-color beads on our confocal (FV1000 with a
> hand-picked 100x/1.4 Plan Super Apo lens), and besides a moderate z-shift=
>
> between blue and other channels (as expected), I have a lateral chromatic=
>
> aberration between the "DAPI" channel (405 nm ex, 430-460 em) and the lon=
> ger
> wavelength channels (green, orange, red fluorescence). So for instance, t=
> he
> colors are colocalized in top left corner of the image, but are several
> pixels off in XY plane in bottom right of the image. The UnwarpJ plugin f=
> or
> ImageJ works well to correct this in individual XY images,

not the same as bUnwarpJ, but similar....

> but 3D correct=
> ion
> may be needed, since the plan correction for the blue channel (405 nm las=
> er
> illumination) is not as good as for the rest of the spectrum.=20=20=20

thats often the case,
as well as the UC laser coming in through a different collimator, so its often
0.5 microns or so off the vis lines.


as i understand on the FV1000 system the 405 and other visible wavelength lasers come through the same fiber and use the same collimator. Check your pinhole alignment and maybe get the laser combiner realigned? Maybe the light coming from the fiber is not centered when it goes through the collimator? What does Olympus suggest?

best wishes

Andreas

[Guy Cox-2]

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

Alan Boyde's TSM work, while fun and interesting, does not accurately show up the CA of the lens when used as it is intended.  Abbe's original apochromat design corrected axial chromatic aberration at 3 wavelengths (and SA at 2).  But it left a lot of lateral chromatic aberration (chromatic difference of magnification).  This was corrected by matching eyepieces (compensating eyepieces) so that one saw a beautifully corrected image.  (With normal eyepieces the result was woeful).  

 

This principle has been followed by most manufacturers ever since, and to avoid the need to swap eyepieces achromat and fluorite lenses were designed to require the same correction (even though they didn't need to be).  When the change was made to infinite tube length some (not all) manufacturers switched the correction to the tube lens.  

 

What has this to do with Boyde's TSM?  In a Petràn type TSM, the spinning disk lies between the eyepieces and the objective, so the pinhole image is not corrected.  This means that in-focus spots at the red and blue ends of the spectrum will miss their respective pinholes.  The resulting colour patterns can be very pretty but are quite complex to interpret.   They do not tell us what the lens will do when used correctly.

 

When I learnt the principles of microscopy (from the legendary JR Baker, 45 years ago) I was introduced to aberration diagrams, which show the exact focal length excursion with wavelength.  Back then these were just a useful figure of merit for lenses.  Now, with advanced 3-D microscopy, they are essential.  But it is very hard to get them out of manufacturers.  I have once - only - seen a manufacturer show one of these curves (and a very impressive one it was) but that was only in a slide - I've never got one on paper.  We really need to pressure manufacturers into giving us this information.

 

                                                                                             Guy

 

Optical Imaging Techniques in Cell Biology

by Guy Cox    CRC Press / Taylor & Francis

     http://www.guycox.com/optical.htm <http://www.guycox.com/optical.htm>

______________________________________________

Associate Professor Guy Cox, MA, DPhil(Oxon)

Australian Centre for Microscopy & Microanalysis,

Madsen Building F09, University of Sydney, NSW 2006

 

Phone [phone]     Fax [phone]

             Mobile [phone]

______________________________________________

      http://www.guycox.net <http://www.guycox.net>

 

 

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of James Pawley
Sent: Saturday, 15 January 2011 4:30 AM
To: [hidden email]
Subject: Re: Objective lens chromatic aberration - Color shift correction for colocalization analysis - was Glycerol Objectives - experience with

 

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

Hi all,

Chromatic aberration is a measure of how the focal length (f) of a
lens changes with wavelength. As the magnification of any optical
system depends on the focal length of the lenses involved, of course
this means that, to the extent that chromatic aberration is present,
the magnification will change with wavelength. In simple optical
systems it is visible as the pronounced red or blue fringes around
the edges of images of high-contrast features. This is called lateral
chromatic aberration and, if the optical system is properly aligned,
it "should" be essentially invisible in the center of the field and
get larger as you approach the edges. (The finding that it lines up
in one corner but not elsewhere makes me think that there may be an
alignment issue.)

As mentioned before all refractive lenses have some chromatic
aberration. (Reflective lenses do not). A good plan-apo, may be
corrected for 4 wavelengths but the wavelengths chosen may vary from
lens to lens. Between these wavelengths there will show
wavelength-dependent shifts in focal length. The manufacturer
endeavors to keep these shifts less than the depth of field of the
objective so that they are not annoying to the naked eye when viewing
multicolored specimens. In practice you will only see the effects
when viewing planar objects such as patterns in thin metal films that
have been evaporated onto the glass surface.

Alan Boyde was the first to point out that these variations in f were
found in even the most expensive objectives, using a tandem-disk
scanner with Hg illumination (which has a number of intensity peaks
in the visible) and making backscattered/reflected light images of
planar features (surface of the slide). In fact he used this
"problem" to make the first truly live-time 3D confocal microscope:
Lenses were designed to have pronounced linear chromatic aberration.
Consequently, different planes in the specimen satisfied the confocal
conditions at different wavelengths. He then used prisms or
diffraction elements in the oculars that were oriented so that, say,
the red image was slightly displaced to the inside (towards the other
eye) of the blue. Looking through the oculars, the viewer then saw a
stereo image, obtained at full NA that was also color-coded for
topographic height. Because confocal images could be obtained at
several focal depths simultaneously, it was no longer necessary to
move the specimen in z to obtain a focal series, making live-stereo
images possible. These livetime stereo images could even be recorded
on tape and shown to groups using special glasses

 From this you can see that any attempt to correct for CA must be able
to respond to changes in magnification with wavelength as well as the
vertical displacements between the z-location of the on-axis peak
confocal signal vs wavelength. As noted above, there may also be
misalignment issues.

Anyone still willing to try to undertake this task should spend a lot
of time studying the work of Cremer and Cremer in the 1980s and
1990s. These scientists demonstrated one of the first effective
systems of super-resolution widefield and confocal imaging as applied
to FISH (fluorescent in-sit hybridization). They labelled one gene on
a chromosome red, and another nearby one green. They then recorded
the green spot and found its centroid. DItto, the red spot. Because
these two spots were separated by wavelength, their images did not
overlap in the recording system and therefore the distance between
them could be measured with an accuracy that depended on the ability
to determine the position of the centroid of each spot, rather than
on the Abbe resolution of the optics. However, as the position of the
centroid did depend on the CA of the optical system, they had to make
very careful measurements of this parameter in order to apply the
proper correction factors.

Hope to see some of you in Vancouver next June!

Cheers,

Jim Pawley

***************************************************************************
Prof. James B. Pawley,                                      Ph.
608-238-3953                              
21. N. Prospect Ave. Madison, WI 53726 USA
[hidden email]
3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/           Applications due by March 15, 2011
               "If it ain't diffraction, it must be statistics." Anon.

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>Hi Dan,
>The translational correction is very useful,  but what if the shift varies
>across the field?
>Do you know if there is a plugin or a script to perform affine transform in
>3D? The goal would be to "warp" one channel in XYZ to match the
>second channel.
>I have imaged 100 nm multi-color beads on our confocal (FV1000 with a
>hand-picked 100x/1.4 Plan Super Apo lens), and besides a moderate z-shift
>between blue and other channels (as expected), I have a lateral chromatic
>aberration between the "DAPI" channel (405 nm ex, 430-460 em) and the longer
>wavelength channels (green, orange, red fluorescence). So for instance, the
>colors are colocalized in top left corner of the image, but are several
>pixels off in XY plane in bottom right of the image. The UnwarpJ plugin for
>ImageJ works well to correct this in individual XY images, but 3D correction
>may be needed, since the plan correction for the blue channel (405 nm laser
>illumination) is not as good as for the rest of the spectrum.
>
>
>
>Stan Vitha
>Microscopy and Imaging Center
>Texas A&M University
>
>
>On Tue, 11 Jan 2011 11:23:02 +0100, Daniel James White
><[hidden email]> wrote:
>
>>*****
>>To join, leave or search the confocal microscopy listserv, go to:
>>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>*****
>>
>>Hi Martin,
>>
>>On Jan 11, 2011, at 7:01 AM, CONFOCALMICROSCOPY automatic digest
>>system wrote:
>>
>>>  Date:    Mon, 10 Jan 2011 17:18:29 -0600
>>>  From:    Martin Wessendorf <[hidden email]>
>>>  Subject: Re: Glycerol Objectives - experience with
>>>
>>>  *****
>>>  To join, leave or search the confocal microscopy listserv, go to:
>>>  http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>>  *****
>>>
>>>  On 1/10/2011 4:32 PM, Rosemary White wrote:
>>>
>>>>  I've found that the red and blue emission aren't quite lined up
>>>>vertically,
>>>>  at least with our "blue" 63x objective, so have to cut the top one or two
>>>>  slices off a blue vertical series (depending on depth of slice), and the
>>>>  bottom slices off a red series and reassemble to get the
>>>>emissions aligned -
>>>>  i.e. from the same depth in the tissue.  The objectives do vary a bit,
>>>>  perhaps ours isn't as well corrected as some.
>>>
>>>  Can anyone explain where the difficulty arises in correcting for axial
>>>  chromatic aberration?  I've consistently seen serious axial chromatic
>>>  aberration in good quality oil objectives from one very reputable
>>>  manufacturer, between red, green and far-red (1 um off in the z-axis
>>>  between green and red, and between red and far-red; 2 um between green
>>>  and far-red).  I have always heard that axial chromatic aberration was
>>>  easy to correct for and would've thought that a solution for 3-color
>>>  correction would've been found 100 years ago.  However, I don't know
>>>  enough optics to understand the subtleties.  Are there trade-offs to
>>>  trying to obtain good correction, besides the cost in scattering of
>>>  adding additional lens elements?
>>
>>Sadly, not all objective lenses were made equal.
>>
>>In our hands on point scanning confocals, even when you align the
>pinhole(s) as best you can
>>there can still be a micron of misalignment in between "green", "red" or
>"far red",
>>with DAPI etc. often being way off, as it comes through a different optic
>fiber and collimator anyway (eg on an LSM 510)
>>
>>I have tested different lenses of the same spec from the same manufacturer,
>>and found that they all have their own personality.
>>Good correction is very possible, but "good" is a relative thing.
>  >
>>I send back ones that are too "bad", and keep the "good" ones.
>>I take z stacks of 1 micron multi colour  beads to measure the
>>remaining error
>>(dont need sub resolution bead images for this measurement)
>>
>>There will always be a significant error, even in the very best
>>"Super-dooper Mega Extra Apo-Chromat"  lens you can get from any
>manufacturer for less than 100 k dollars.
>>They simply can not be made perfect at a reasonable cost (as explained to
>me by a Zeiss lens guru)
>>
>>If you want to do precise colocalization studies at the highest optical
>resolution (everyone does, even if they don't immediately realize it)
>>then you simply MUST measure the error, then correct/shift the images (in
>3D) before colocalization/correlation analysis.
>>
>>This can be done in 3D in ImageJ/ Fiji
>>using nice interpolation methods from Erik Meijering, for sub pixel shifts
>- TJ-Translate :
>>http://pacific.mpi-cbg.de/wiki/index.php/TransformJ
>>leading to
>>http://imagescience.org/meijering/software/transformj/translate.html
>>
>>or in the great colour shift corrector of Huygens Professional (no
>commercial interest - just a happy customer)
>>http://www.svi.nl/ChromaticShiftCorrector
>>
>>or roughly in the Zeiss AIM 510 software in xy only (whole pixel shifts),
>>
>>The effect on the 2 channel scatterplot / 2D histogram / fluorogram is very
>significant.
>>A ugly poorly correlated cloud turns into tight correlated populations of
>pixels,
>>and the coloc. coefficients jump much higher.
>>
>>more info here:
>>http://ifn.mpi-cbg.de/wiki/ifn/index.php/Chromatic_aberration_measurement_and_correction
>>
>>cheers
>>
>>Dan
>>
>>
>>
>>>
>>>  Thanks--
>>>
>>>  Martin
>>>  --
>>>  Martin Wessendorf, Ph.D.                   office: [phone]
>>>  Assoc Prof, Dept Neuroscience                 lab: [phone]
>>>  University of Minnesota             Preferred FAX: [phone]
>>>  6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
>>>  Minneapolis, MN  55455                    e-mail: [hidden email]
>>
>>Dr. Daniel James White BSc. (Hons.) PhD
>>Senior Microscopist / Image Visualisation, Processing and Analysis
>>Light Microscopy and Image Processing Facilities
>>Max Planck Institute of Molecular Cell Biology and Genetics
>>Pfotenhauerstrasse 108
>>01307 DRESDEN
>>Germany
>>
>>[phone] (German Mobile)
>>[phone] (Work phone at MPI-CBG)
>>[phone] (Fax MPI-CBG LMF)
>>
>>http://www.bioimagexd.net     BioImageXD
>>http://pacific.mpi-cbg.de             Fiji -  is just ImageJ
>>(Batteries Included)
>>http://www.chalkie.org.uk             Dan's Homepages
>>https://ifn.mpi-cbg.de                        Dresden Imaging
>>Facility Network
>>dan (at) chalkie.org.uk
>>( white (at) mpi-cbg.de )

--
***************************************************************************
Prof. James B. Pawley,                                      Ph.
608-238-3953                              
21. N. Prospect Ave. Madison, WI 53726 USA
[hidden email]
3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/           Applications due by March 15, 2011
               "If it ain't diffraction, it must be statistics." Anon.

________________________________

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1191 / Virus Database: 1435/3342 - Release Date: 12/27/10
Internal Virus Database is out of date.

[George McNamara]

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

Hi Stan,

Plucking one idea from Jim Pawley's post - the objective lens should be
very well corrected in the center of the field - zoom up to 2x (or more)
and use motorized scanning stage tiling if you need a larger field of
view. If the stitching software does not not do seamless stitching,
complain to the manufacturer until they fix it.

Sincerely,

George
p.s. For 2D stitching, Adobe Photoshop CS5 (and 3 and 4) have a very
nice multi-file stitching command (File menu - Automate, or something
like that). Carl Boswell and/or Jerry Sedgewick had mentioned this to me
several years ago (CS3), but I only recently upgraded from 2 to 5.

On 1/13/2011 4:28 PM, Stanislav Vitha wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi Dan,
> The translational correction is very useful,  but what if the shift varies
> across the field?
> Do you know if there is a plugin or a script to perform affine transform in
> 3D? The goal would be to "warp" one channel in XYZ to match the second channel.
> I have imaged 100 nm multi-color beads on our confocal (FV1000 with a
> hand-picked 100x/1.4 Plan Super Apo lens), and besides a moderate z-shift
> between blue and other channels (as expected), I have a lateral chromatic
> aberration between the "DAPI" channel (405 nm ex, 430-460 em) and the longer
> wavelength channels (green, orange, red fluorescence). So for instance, the
> colors are colocalized in top left corner of the image, but are several
> pixels off in XY plane in bottom right of the image. The UnwarpJ plugin for
> ImageJ works well to correct this in individual XY images, but 3D correction
> may be needed, since the plan correction for the blue channel (405 nm laser
> illumination) is not as good as for the rest of the spectrum.
>
>
>
> Stan Vitha
> Microscopy and Imaging Center
> Texas A&M University
>
>
> On Tue, 11 Jan 2011 11:23:02 +0100, Daniel James White<[hidden email]>  wrote:
>
>    
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>>
>> Hi Martin,
>>
>> On Jan 11, 2011, at 7:01 AM, CONFOCALMICROSCOPY automatic digest system wrote:
>>
>>      
>>> Date:    Mon, 10 Jan 2011 17:18:29 -0600
>>> From:    Martin Wessendorf<[hidden email]>
>>> Subject: Re: Glycerol Objectives - experience with
>>>
>>> *****
>>> To join, leave or search the confocal microscopy listserv, go to:
>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>> *****
>>>
>>> On 1/10/2011 4:32 PM, Rosemary White wrote:
>>>
>>>        
>>>> I've found that the red and blue emission aren't quite lined up vertically,
>>>> at least with our "blue" 63x objective, so have to cut the top one or two
>>>> slices off a blue vertical series (depending on depth of slice), and the
>>>> bottom slices off a red series and reassemble to get the emissions aligned -
>>>> i.e. from the same depth in the tissue.  The objectives do vary a bit,
>>>> perhaps ours isn't as well corrected as some.
>>>>          
>>> Can anyone explain where the difficulty arises in correcting for axial
>>> chromatic aberration?  I've consistently seen serious axial chromatic
>>> aberration in good quality oil objectives from one very reputable
>>> manufacturer, between red, green and far-red (1 um off in the z-axis
>>> between green and red, and between red and far-red; 2 um between green
>>> and far-red).  I have always heard that axial chromatic aberration was
>>> easy to correct for and would've thought that a solution for 3-color
>>> correction would've been found 100 years ago.  However, I don't know
>>> enough optics to understand the subtleties.  Are there trade-offs to
>>> trying to obtain good correction, besides the cost in scattering of
>>> adding additional lens elements?
>>>        
>> Sadly, not all objective lenses were made equal.
>>
>> In our hands on point scanning confocals, even when you align the
>>      
> pinhole(s) as best you can
>    
>> there can still be a micron of misalignment in between "green", "red" or
>>      
> "far red",
>    
>> with DAPI etc. often being way off, as it comes through a different optic
>>      
> fiber and collimator anyway (eg on an LSM 510)
>    
>> I have tested different lenses of the same spec from the same manufacturer,
>> and found that they all have their own personality.
>> Good correction is very possible, but "good" is a relative thing.
>>
>> I send back ones that are too "bad", and keep the "good" ones.
>> I take z stacks of 1 micron multi colour  beads to measure the remaining error
>> (dont need sub resolution bead images for this measurement)
>>
>> There will always be a significant error, even in the very best
>> "Super-dooper Mega Extra Apo-Chromat"  lens you can get from any
>>      
> manufacturer for less than 100 k dollars.
>    
>> They simply can not be made perfect at a reasonable cost (as explained to
>>      
> me by a Zeiss lens guru)
>    
>> If you want to do precise colocalization studies at the highest optical
>>      
> resolution (everyone does, even if they don't immediately realize it)
>    
>> then you simply MUST measure the error, then correct/shift the images (in
>>      
> 3D) before colocalization/correlation analysis.
>    
>> This can be done in 3D in ImageJ/ Fiji
>> using nice interpolation methods from Erik Meijering, for sub pixel shifts
>>      
> - TJ-Translate :
>    
>> http://pacific.mpi-cbg.de/wiki/index.php/TransformJ
>> leading to
>> http://imagescience.org/meijering/software/transformj/translate.html
>>
>> or in the great colour shift corrector of Huygens Professional (no
>>      
> commercial interest - just a happy customer)
>    
>> http://www.svi.nl/ChromaticShiftCorrector
>>
>> or roughly in the Zeiss AIM 510 software in xy only (whole pixel shifts),
>>
>> The effect on the 2 channel scatterplot / 2D histogram / fluorogram is very
>>      
> significant.
>    
>> A ugly poorly correlated cloud turns into tight correlated populations of
>>      
> pixels,
>    
>> and the coloc. coefficients jump much higher.
>>
>> more info here:
>> http://ifn.mpi-cbg.de/wiki/ifn/index.php/Chromatic_aberration_measurement_and_correction
>>
>> cheers
>>
>> Dan
>>
>>
>>
>>      
>>> Thanks--
>>>
>>> Martin
>>> --
>>> Martin Wessendorf, Ph.D.                   office: [phone]
>>> Assoc Prof, Dept Neuroscience                 lab: [phone]
>>> University of Minnesota             Preferred FAX: [phone]
>>> 6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
>>> Minneapolis, MN  55455                    e-mail: [hidden email]
>>>        
>> Dr. Daniel James White BSc. (Hons.) PhD
>> Senior Microscopist / Image Visualisation, Processing and Analysis
>> Light Microscopy and Image Processing Facilities
>> Max Planck Institute of Molecular Cell Biology and Genetics
>> Pfotenhauerstrasse 108
>> 01307 DRESDEN
>> Germany
>>
>> [phone] (German Mobile)
>> [phone] (Work phone at MPI-CBG)
>> [phone] (Fax MPI-CBG LMF)
>>
>> http://www.bioimagexd.net  BioImageXD
>> http://pacific.mpi-cbg.de                Fiji -  is just ImageJ (Batteries Included)
>> http://www.chalkie.org.uk                Dan's Homepages
>> https://ifn.mpi-cbg.de  Dresden Imaging Facility Network
>> dan (at) chalkie.org.uk
>> ( white (at) mpi-cbg.de )
>>      
>    

[James Pawley]

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

Hi again,

Not really disagreeing with Guy, but he makes my
point. Yes, when imaging a point object that
emitted both red and blue light using a lens with
CA (such as one where the residual CA was
supposed to be corrected in the ocular), either
the red or the green would be properly focused at
the plane of the spinning disk (but not both!).

However, when the whitish (Hg) light starts from
the disk, and goes through the CA objective, each
color does form a spot at some focal distance,
creating a series of spots at different focus
planes (for different colors) in the specimen.
When light is reflected or scattered from a point
object located, say, where the red light focuses,
this scattered or reflected light will be in
focus when it gets back to the disk (or on the
Petran disk, to the opposite side of the
symmetrical disk). Likewise, a point scatterer
located where the blue light forms a focus will
also be in focus when it reaches the disk plane.
This is how the color of the light that makes it
through the disk (and therefore looks "bright")
codes for specimen topography. In this case you
would not want to use a compensating eyepiece as
you would want to convey the disk image to the
recording plane.

I also agree that, optically speaking, one should
not interpose a scanning disk into the
intermediate image plane of any optical system in
which, CA errors in the objective are designed to
be corrected in the ocular. And that using such
lenses was one of the reasons for the color
effects in early BSL imaging using disk-scaners.
This is the reason that most such systems use
Nikon CF and later CFI objectives in which CA was
corrected in both the objective and the occular.

To make color code unambiguously for height,
Boyde convinced Leningrad Optical to make some
special "linear longitudinal chromatic
dispersion" objectives in which the focal length
varied inversely with the wavelength (i.e, not
achromats or apochromats, just simple uncorrected
lenses) so that each focal depth would match the
confocal condition (i.e., be in focus) for only a
single color. Although this produced stunning 3D
confocal movies, it did still have the lateral CA
problem in that blue features would be magnified
a bit more than red ones, but this was not too
serious if you stuck to the center of the field
of view.

Regarding the early Abbe objectives, it is true
that, accompanied with the correct eyepiece, they
were as good as Guy says. However they also had
significant curvature of field and therefore they
were less successful than modern objectives when
viewing large planar specimens.

Boyde's system is described in some detail in
Chapter 15 of the Second Edition of the Handbook.

Cheers,

Jim Pawley

>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>Alan Boyde's TSM work, while fun and
>interesting, does not accurately show up the CA
>of the lens when used as it is intended.  Abbe's
>original apochromat design corrected axial
>chromatic aberration at 3 wavelengths (and SA at
>2).  But it left a lot of lateral chromatic
>aberration (chromatic difference of
>magnification).  This was corrected by matching
>eyepieces (compensating eyepieces) so that one
>saw a beautifully corrected image.  (With normal
>eyepieces the result was woeful).
>
>
>
>This principle has been followed by most
>manufacturers ever since, and to avoid the need
>to swap eyepieces achromat and fluorite lenses
>were designed to require the same correction
>(even though they didn't need to be).  When the
>change was made to infinite tube length some
>(not all) manufacturers switched the correction
>to the tube lens.
>
>
>
>What has this to do with Boyde's TSM?  In a
>Petràn type TSM, the spinning disk lies between
>the eyepieces and the objective, so the pinhole
>image is not corrected.  This means that
>in-focus spots at the red and blue ends of the
>spectrum will miss their respective pinholes.
>The resulting colour patterns can be very pretty
>but are quite complex to interpret.   They do
>not tell us what the lens will do when used
>correctly.
>
>
>
>When I learnt the principles of microscopy (from
>the legendary JR Baker, 45 years ago) I was
>introduced to aberration diagrams, which show
>the exact focal length excursion with
>wavelength.  Back then these were just a useful
>figure of merit for lenses.  Now, with advanced
>3-D microscopy, they are essential.  But it is
>very hard to get them out of manufacturers.  I
>have once - only - seen a manufacturer show one
>of these curves (and a very impressive one it
>was) but that was only in a slide - I've never
>got one on paper.  We really need to pressure
>manufacturers into giving us this information.
>
>
>
>
>Guy
>
>
>
>Optical Imaging Techniques in Cell Biology
>
>by Guy Cox    CRC Press / Taylor & Francis
>
>      http://www.guycox.com/optical.htm <http://www.guycox.com/optical.htm>
>
>______________________________________________
>
>Associate Professor Guy Cox, MA, DPhil(Oxon)
>
>Australian Centre for Microscopy & Microanalysis,
>
>Madsen Building F09, University of Sydney, NSW 2006
>
>
>
>Phone [phone]     Fax [phone]
>
>              Mobile [phone]
>
>______________________________________________
>
>       http://www.guycox.net <http://www.guycox.net>
>
>
>
>
>
>From: Confocal Microscopy List
>[mailto:[hidden email]] On
>Behalf Of James Pawley
>Sent: Saturday, 15 January 2011 4:30 AM
>To: [hidden email]
>Subject: Re: Objective lens chromatic aberration
>- Color shift correction for colocalization
>analysis - was Glycerol Objectives - experience
>with
>
>
>
>*****
>To join, leave or search the confocal microscopy listserv, go to:
>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>*****
>
>Hi all,
>
>Chromatic aberration is a measure of how the focal length (f) of a
>lens changes with wavelength. As the magnification of any optical
>system depends on the focal length of the lenses involved, of course
>this means that, to the extent that chromatic aberration is present,
>the magnification will change with wavelength. In simple optical
>systems it is visible as the pronounced red or blue fringes around
>the edges of images of high-contrast features. This is called lateral
>chromatic aberration and, if the optical system is properly aligned,
>it "should" be essentially invisible in the center of the field and
>get larger as you approach the edges. (The finding that it lines up
>in one corner but not elsewhere makes me think that there may be an
>alignment issue.)
>
>As mentioned before all refractive lenses have some chromatic
>aberration. (Reflective lenses do not). A good plan-apo, may be
>corrected for 4 wavelengths but the wavelengths chosen may vary from
>lens to lens. Between these wavelengths there will show
>wavelength-dependent shifts in focal length. The manufacturer
>endeavors to keep these shifts less than the depth of field of the
>objective so that they are not annoying to the naked eye when viewing
>multicolored specimens. In practice you will only see the effects
>when viewing planar objects such as patterns in thin metal films that
>have been evaporated onto the glass surface.
>
>Alan Boyde was the first to point out that these variations in f were
>found in even the most expensive objectives, using a tandem-disk
>scanner with Hg illumination (which has a number of intensity peaks
>in the visible) and making backscattered/reflected light images of
>planar features (surface of the slide). In fact he used this
>"problem" to make the first truly live-time 3D confocal microscope:
>Lenses were designed to have pronounced linear chromatic aberration.
>Consequently, different planes in the specimen satisfied the confocal
>conditions at different wavelengths. He then used prisms or
>diffraction elements in the oculars that were oriented so that, say,
>the red image was slightly displaced to the inside (towards the other
>eye) of the blue. Looking through the oculars, the viewer then saw a
>stereo image, obtained at full NA that was also color-coded for
>topographic height. Because confocal images could be obtained at
>several focal depths simultaneously, it was no longer necessary to
>move the specimen in z to obtain a focal series, making live-stereo
>images possible. These livetime stereo images could even be recorded
>on tape and shown to groups using special glasses
>
>  From this you can see that any attempt to correct for CA must be able
>to respond to changes in magnification with wavelength as well as the
>vertical displacements between the z-location of the on-axis peak
>confocal signal vs wavelength. As noted above, there may also be
>misalignment issues.
>
>Anyone still willing to try to undertake this task should spend a lot
>of time studying the work of Cremer and Cremer in the 1980s and
>1990s. These scientists demonstrated one of the first effective
>systems of super-resolution widefield and confocal imaging as applied
>to FISH (fluorescent in-sit hybridization). They labelled one gene on
>a chromosome red, and another nearby one green. They then recorded
>the green spot and found its centroid. DItto, the red spot. Because
>these two spots were separated by wavelength, their images did not
>overlap in the recording system and therefore the distance between
>them could be measured with an accuracy that depended on the ability
>to determine the position of the centroid of each spot, rather than
>on the Abbe resolution of the optics. However, as the position of the
>centroid did depend on the CA of the optical system, they had to make
>very careful measurements of this parameter in order to apply the
>proper correction factors.
>
>Hope to see some of you in Vancouver next June!
>
>Cheers,
>
>Jim Pawley
>
>***************************************************************************
>Prof. James B. Pawley,                                      Ph.
>608-238-3953
>21. N. Prospect Ave. Madison, WI 53726 USA
>[hidden email]
>3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada
>Info: http://www.3dcourse.ubc.ca/           Applications due by March 15, 2011
>                "If it ain't diffraction, it must be statistics." Anon.
>
>>*****
>>To join, leave or search the confocal microscopy listserv, go to:
>>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>*****
>>
>>Hi Dan,
>>The translational correction is very useful,  but what if the shift varies
>>across the field?
>>Do you know if there is a plugin or a script to perform affine transform in
>>3D? The goal would be to "warp" one channel in XYZ to match the
>>second channel.
>>I have imaged 100 nm multi-color beads on our confocal (FV1000 with a
>>hand-picked 100x/1.4 Plan Super Apo lens), and besides a moderate z-shift
>>between blue and other channels (as expected), I have a lateral chromatic
>>aberration between the "DAPI" channel (405 nm ex, 430-460 em) and the longer
>>wavelength channels (green, orange, red fluorescence). So for instance, the
>>colors are colocalized in top left corner of the image, but are several
>>pixels off in XY plane in bottom right of the image. The UnwarpJ plugin for
>>ImageJ works well to correct this in individual XY images, but 3D correction
>>may be needed, since the plan correction for the blue channel (405 nm laser
>>illumination) is not as good as for the rest of the spectrum.
>>
>>
>>
>>Stan Vitha
>>Microscopy and Imaging Center
>>Texas A&M University
>>
>>
>>On Tue, 11 Jan 2011 11:23:02 +0100, Daniel James White
>><[hidden email]> wrote:
>>
>>>*****
>>>To join, leave or search the confocal microscopy listserv, go to:
>>>http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>>*****
>>>
>>>Hi Martin,
>>>
>>>On Jan 11, 2011, at 7:01 AM, CONFOCALMICROSCOPY automatic digest
>>>system wrote:
>>>
>>>>   Date:    Mon, 10 Jan 2011 17:18:29 -0600
>>>>   From:    Martin Wessendorf <[hidden email]>
>>>>   Subject: Re: Glycerol Objectives - experience with
>>>>
>>>>   *****
>>>>   To join, leave or search the confocal microscopy listserv, go to:
>>>>   http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>>>   *****
>>>>
>>>>   On 1/10/2011 4:32 PM, Rosemary White wrote:
>>>>
>>>>>   I've found that the red and blue emission aren't quite lined up
>>>>>vertically,
>>>>>   at least with our "blue" 63x objective, so
>>>>>have to cut the top one or two
>>>>>   slices off a blue vertical series (depending on depth of slice), and the
>  >>>>  bottom slices off a red series and reassemble to get the
>>>>>emissions aligned -
>>>>>   i.e. from the same depth in the tissue.  The objectives do vary a bit,
>>>>>   perhaps ours isn't as well corrected as some.
>>>>
>>>>   Can anyone explain where the difficulty arises in correcting for axial
>>>>   chromatic aberration?  I've consistently seen serious axial chromatic
>>>>   aberration in good quality oil objectives from one very reputable
>>>>   manufacturer, between red, green and far-red (1 um off in the z-axis
>>>>   between green and red, and between red and far-red; 2 um between green
>>>>   and far-red).  I have always heard that axial chromatic aberration was
>>>>   easy to correct for and would've thought that a solution for 3-color
>>>>   correction would've been found 100 years ago.  However, I don't know
>>>>   enough optics to understand the subtleties.  Are there trade-offs to
>>>>   trying to obtain good correction, besides the cost in scattering of
>>>>   adding additional lens elements?
>>>
>>>Sadly, not all objective lenses were made equal.
>>>
>>>In our hands on point scanning confocals, even when you align the
>>pinhole(s) as best you can
>>>there can still be a micron of misalignment in between "green", "red" or
>>"far red",
>>>with DAPI etc. often being way off, as it comes through a different optic
>  >fiber and collimator anyway (eg on an LSM 510)
>>>
>>>I have tested different lenses of the same spec from the same manufacturer,
>>>and found that they all have their own personality.
>>>Good correction is very possible, but "good" is a relative thing.
>>   >
>>>I send back ones that are too "bad", and keep the "good" ones.
>>>I take z stacks of 1 micron multi colour  beads to measure the
>>>remaining error
>>>(dont need sub resolution bead images for this measurement)
>>>
>>>There will always be a significant error, even in the very best
>>>"Super-dooper Mega Extra Apo-Chromat"  lens you can get from any
>>manufacturer for less than 100 k dollars.
>>>They simply can not be made perfect at a reasonable cost (as explained to
>>me by a Zeiss lens guru)
>>>
>>>If you want to do precise colocalization studies at the highest optical
>>resolution (everyone does, even if they don't immediately realize it)
>>>then you simply MUST measure the error, then correct/shift the images (in
>>3D) before colocalization/correlation analysis.
>>>
>>>This can be done in 3D in ImageJ/ Fiji
>>>using nice interpolation methods from Erik Meijering, for sub pixel shifts
>>- TJ-Translate :
>>>http://pacific.mpi-cbg.de/wiki/index.php/TransformJ
>>>leading to
>>>http://imagescience.org/meijering/software/transformj/translate.html
>>>
>>>or in the great colour shift corrector of Huygens Professional (no
>>commercial interest - just a happy customer)
>>>http://www.svi.nl/ChromaticShiftCorrector
>>>
>>>or roughly in the Zeiss AIM 510 software in xy only (whole pixel shifts),
>>>
>>>The effect on the 2 channel scatterplot / 2D histogram / fluorogram is very
>>significant.
>>>A ugly poorly correlated cloud turns into tight correlated populations of
>>pixels,
>>>and the coloc. coefficients jump much higher.
>>>
>>>more info here:
>>>http://ifn.mpi-cbg.de/wiki/ifn/index.php/Chromatic_aberration_measurement_and_correction
>>>
>>>cheers
>>>
>>>Dan
>>>
>>>
>>>
>>>>
>>>>   Thanks--
>>>>
>>>>   Martin
>>>>   --
>>>>   Martin Wessendorf, Ph.D.                   office: [phone]
>>>>   Assoc Prof, Dept Neuroscience                 lab: [phone]
>>>>   University of Minnesota             Preferred FAX: [phone]
>>>>   6-145 Jackson Hall, 321 Church St. SE    Dept Fax: [phone]
>>>>   Minneapolis, MN  55455                    e-mail: [hidden email]
>>>
>>>Dr. Daniel James White BSc. (Hons.) PhD
>>>Senior Microscopist / Image Visualisation, Processing and Analysis
>>>Light Microscopy and Image Processing Facilities
>>>Max Planck Institute of Molecular Cell Biology and Genetics
>>>Pfotenhauerstrasse 108
>>>01307 DRESDEN
>>>Germany
>>>
>>>[phone] (German Mobile)
>>>[phone] (Work phone at MPI-CBG)
>>>[phone] (Fax MPI-CBG LMF)
>>>
>>>http://www.bioimagexd.net     BioImageXD
>>>http://pacific.mpi-cbg.de             Fiji -  is just ImageJ
>  >>(Batteries Included)
>>>http://www.chalkie.org.uk             Dan's Homepages
>>>https://ifn.mpi-cbg.de                        Dresden Imaging
>>>Facility Network
>>>dan (at) chalkie.org.uk
>>>( white (at) mpi-cbg.de )
>
>
>--
>***************************************************************************
>Prof. James B. Pawley,                                      Ph.
>608-238-3953
>21. N. Prospect Ave. Madison, WI 53726 USA
>[hidden email]
>3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada
>Info: http://www.3dcourse.ubc.ca/           Applications due by March 15, 2011
>                "If it ain't diffraction, it must be statistics." Anon.
>
>________________________________
>
>No virus found in this message.
>Checked by AVG - www.avg.com
>Version: 10.0.1191 / Virus Database: 1435/3342 - Release Date: 12/27/10
>Internal Virus Database is out of date.

--
***************************************************************************
Prof. James B. Pawley,                
Ph.  608-238-3953              
21. N. Prospect Ave. Madison, WI 53726 USA
[hidden email]
3D Microscopy of Living Cells Course, June 11-23, 2011, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/            Applications due by March 15, 2011
               "If it ain't diffraction, it must be statistics." Anon.