Deconvolution of 3D SIM data

[A. Hamdoun]

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

Hi Everyone,

We are starting to process through  OMX data sets and we are writing for any
thoughts on the use of deconvolution in these images. We have heard mixed
things about whether or not it is valid...  Has anyone here deconvolved 3D SIM
data?  Were actual or theoretical PSFs used?

Much thanks,
A

[David Baddeley]

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

I believe that SIM data is usually at least partly deconvolved already in the
reconstruction process, although the extent of the deconvolution depends on the
heritage of your SIM technique. I think all techniques involve at least a Weiner
filtering step, and some might include a iterative optimisation as well. I'd
thus be pretty cautious about performing any further deconvolution. If you do,
you'd want to use very small beads (<50 nm) for your PSF measurements, which
might make getting reliable (ie not noisy) PSF estimates difficult. Its also
quite likely that the factors which end up limiting the resolution will relate
to uneven bleaching ('burn in' of the mask) and mask errors (alignment errors
and sample abberations which change the shape of the illumination pattern from
what is expected). Deconvolution isn't going to help you here.

cheers,
David


----- Original Message ----
To: [hidden email]
Sent: Fri, 8 April, 2011 8:16:08 AM
Subject: Deconvolution of 3D SIM data

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

Hi Everyone,

We are starting to process through  OMX data sets and we are writing for any
thoughts on the use of deconvolution in these images. We have heard mixed
things about whether or not it is valid...  Has anyone here deconvolved 3D SIM
data?  Were actual or theoretical PSFs used?

Much thanks,
A

[Daniel James White]

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

Hi Amro and David,

> Date:    Thu, 7 Apr 2011 13:43:45 -0700
> From:    David Baddeley <[hidden email]>
> Subject: Re: Deconvolution of 3D SIM data
>
>
>> Date:    Thu, 7 Apr 2011 15:16:08 -0500
>> From:    Amro Hamdoun <[hidden email]>
>> Subject: Deconvolution of 3D SIM data
>>
>>
>> Hi Everyone,
>>
>> We are starting to process through  OMX data sets and we are writing for =
>> any=20
>> thoughts on the use of deconvolution in these images.

We got our OMX installed early this year..... this is what I have learned so far.....

in the Applied precision softworx software, the tool that does the SIM reconstruction
takes a measured SIM PSF and uses that to deconvolve the images,
as part of the super resolution reconstruction.

Whichever engineer or application specialist from API that installed your OMX
should have trained at least a couple of folks at your site on all this stuff.
Maybe that was not you.

Anyone who uses this kind of complex system  needs to understand its optics
and the ideas behind the maths that do the reconstruction ( = modeling of the fluorophore distribution)
be it standard widefield deconvolution (eg DeltaVision), normal structured illumination (apotome / optiogrid) , super resolution structured illumination (OMX), PALM/STORM ... etc.

Without that understanding, you have no hope of sensibly interpreting the output images,
and deciding if the image contains bad artifacts or not.

Applied Precision run advanced user training sessions,
and if you want to learn more about what is going on, you might like to arrange to go to Seattle and take part in one.

Failing that, you should read the original papers by Matts Gustaffson that describe the methods.
http://www.hhmi.org/research/groupleaders/gustafsson_bio.html


>> We have heard mixed=
>> =20
>> things about whether or not it is valid...  Has anyone here deconvolved 3=
>> D SIM=20
>> data?  Were actual or theoretical PSFs used?=20

the OMX software provided by API for 3D SIM super resolution reconstruction,
does deconvolution with a measured SIM PSF (not the same as the wide field one)
as part of the image reconstruction.

thus it makes no sense to do an extra deconvoution afterwards...
the result images ia already deconvolved: its a most likely model of the fluorophore distribution,
resolution limited (spatial frequency band limited)  by the physics of the system.

>>
>> Much thanks,
>> A
>
> I believe that SIM data is usually at least partly deconvolved already in t=
> he =0Areconstruction process, although the extent of the deconvolution depe=
> nds on the =0Aheritage of your SIM technique.

OMX software does a full deconvolution, as far as I understand.

> I think all techniques involv=
> e at least a Weiner =0Afiltering step, and some might include a iterative o=
> ptimisation as well. I'd =0Athus be pretty cautious about performing any fu=
> rther deconvolution.

Me too.

> If you do, =0Ayou'd want to use very small beads (<50 =
> nm) for your PSF measurements, which =0Amight make getting reliable (ie not=
> noisy) PSF estimates difficult.

The OMX software needs measured SIM PSF image (one per camera/channel) for the reconstruction...
the engineer that installed it should have made these,
and one should probably make them again every now and then to test system performance.

> Its also =0Aquite likely that the factors =
> which end up limiting the resolution will relate =0Ato uneven bleaching ('b=
> urn in' of the mask) and mask errors (alignment errors =0Aand sample abbera=
> tions which change the shape of the illumination pattern from =0Awhat is ex=
> pected). Deconvolution isn't going to help you here.=0A=0Acheers,=0ADavid=

David is one the money here...

but you really should talk to someone like Paul Goodwin
or one of the application specialists for OMX at at API.

cheers

Dan

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 )

[Hans]

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

(*commercial  response*)

Dear All,

As the previous responses by David and Dan already state very well,
caution should be taken in straightforward applying deconvolution on top
of reconstructed images from SIM systems since these might be the result
of various non-linear processing steps.

Still, since it seems that there is room for improvement in at least
some SIM reconstructions, and the current discussion also points in that
direction, here at SVI we wonder about the need for a SIM deconvolution
module. This would work directly on raw SIM data.
We would very much appreciate to get some off-listfeedback from
(potential) SIM users for such a module.

Cheers, Hans


--
-------------------------------------------------------------------
dr. Hans T.M. van der Voort                           ([hidden email])
Scientific Volume Imaging b.v.,             URL: http://www.svi.nl/

[Andreas Bruckbauer]

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


 When i generate a PSF for deconvolution using suitable beads, then image the same beads again and deconvolve the image, i would expect to get really tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample 
I would expect something similar for the OMX when the reconstruction is perfect and includes proper deconvolution. However we get larger features, still within the expected OMX resolution 120 nm in width and 300 nm in z, but no dots. I think there is still some improvement possible either in the setup of the instrument or the software.

best wishes

Andreas


 

 

-----Original Message-----
From: Hans <[hidden email]>
To: [hidden email]
Sent: Fri, 8 Apr 2011 10:46
Subject: Re: Deconvolution of 3D SIM data


*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy 
*****
 
(*commercial  response*)
 
Dear All,
 
As the previous responses by David and Dan already state very well, caution should be taken in straightforward applying deconvolution on top of reconstructed images from SIM systems since these might be the result of various non-linear processing steps.
 
Still, since it seems that there is room for improvement in at least some SIM reconstructions, and the current discussion also points in that direction, here at SVI we wonder about the need for a SIM deconvolution module. This would work directly on raw SIM data.
We would very much appreciate to get some off-listfeedback from (potential) SIM users for such a module.
 
Cheers, Hans
 
-- -------------------------------------------------------------------
dr. Hans T.M. van der Voort                           ([hidden email])
Scientific Volume Imaging b.v.,             URL: http://www.svi.nl/ 

 

[Northan, Brian]

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

Commercial response from Media Cybernetics:

One thing to note is that depending on the sampling rate you may not get dots back even after deconvolution.  

When interpreting OMX deconvolution results it may be useful to consider the differences between deconvolving widefield and confocal images.

In widefield the dominant effect of deconvolution is deblurring, thus the dramatic haze reduction.  In confocal/two-photon/ the effect is a little different.  The hardware has all ready done some deblurring at the expense of noise.  So in these modalities the effect of deconvolution is some deblurring combined with noise reduction.  

I have deconvolved some SI images from Optigrid and the Quorum Angstrom using Autoquant blind deconvolution.  As when deconvolving confocal the result was some deblurring and some noise reduction resulting in an increased signal to noise.  

So the expectation when deconvolving OMX is not necessarily that you get exactly a "dot" back (especially if the image is over-sampled) but to get increased contrast and less noise.

As the other posters have mentioned the situation is complicated by possible non-linear processing steps in SIM. And care must be taken when using deconvolution on a modality it is not optimized for.  

We encourage anyone who is interested in deconvolving SIM images to give us feedback.

Thanks

Brian


-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Andreas Bruckbauer
Sent: Friday, April 08, 2011 11:49 AM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

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


 When i generate a PSF for deconvolution using suitable beads, then image the same beads again and deconvolve the image, i would expect to get really tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample 
I would expect something similar for the OMX when the reconstruction is perfect and includes proper deconvolution. However we get larger features, still within the expected OMX resolution 120 nm in width and 300 nm in z, but no dots. I think there is still some improvement possible either in the setup of the instrument or the software.

best wishes

Andreas


 

 

-----Original Message-----
From: Hans <[hidden email]>
To: [hidden email]
Sent: Fri, 8 Apr 2011 10:46
Subject: Re: Deconvolution of 3D SIM data


*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy 
*****
 
(*commercial  response*)
 
Dear All,
 
As the previous responses by David and Dan already state very well, caution should be taken in straightforward applying deconvolution on top of reconstructed images from SIM systems since these might be the result of various non-linear processing steps.
 
Still, since it seems that there is room for improvement in at least some SIM reconstructions, and the current discussion also points in that direction, here at SVI we wonder about the need for a SIM deconvolution module. This would work directly on raw SIM data.
We would very much appreciate to get some off-listfeedback from (potential) SIM users for such a module.
 
Cheers, Hans
 
-- -------------------------------------------------------------------
dr. Hans T.M. van der Voort                           ([hidden email])
Scientific Volume Imaging b.v.,             URL: http://www.svi.nl/ 

 
######################################################################################
CONFIDENTIALITY NOTICE:
This email transmission and its attachments contain confidential and proprietary information
of Princeton Instruments, Acton Research, Media Cybernetics and their affiliates and is
intended for the exclusive and confidential use of the intended recipient. Any use, dissemination,
printing, or copying of this transmission and its attachment(s) is strictly prohibited. If you
are not the intended recipient, please do not read, print, copy, distribute or take action in
reliance upon this message.  If you have received this in error, please notify the sender immediately
by telephone or return email and promptly delete all copies of the original transmission and its
attachments from your computer system.
#######################################################################################

[Gert van Cappellen-2]

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

The problem with reconstruction of the SIM images is I think as with
most other imaging techniques the signal to noise ratio of the raw
images. An optimal deconvolution should there start on the raw images
and not on the reconstructed images. As stated before all SIM algorithms
do already contain some deconvolution, but I  think there is room for
improvement. With a better knowledge of the deformations of the grid in
the images and the effect in biological specimen the reconstruction
based on deconvolution might improve. This needs a lot of testing with
different biological samples or other tricks.

With kind regards,

Gert van Cappellen
Optical Imaging Centre
Erasmus MC (Erasmus Medical Centre)
Rotterdam
Netherlands


Op 7-4-2011 22:16, Amro Hamdoun schreef:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hi Everyone,
>
> We are starting to process through  OMX data sets and we are writing for any
> thoughts on the use of deconvolution in these images. We have heard mixed
> things about whether or not it is valid...  Has anyone here deconvolved 3D SIM
> data?  Were actual or theoretical PSFs used?
>
> Much thanks,
> A

[Daniel James White]

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

Hi Andreas,

On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system wrote:

> Date:    Fri, 8 Apr 2011 11:48:46 -0400
> From:    Andreas Bruckbauer <[hidden email]>
> Subject: Re: Deconvolution of 3D SIM data
>
>
> When i generate a PSF for deconvolution using suitable beads, then image=
> the same beads again and deconvolve the image, i would expect to get real=
> ly tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample=20

real tiny? The images of sub resolution objects dont get very much smaller after deconvolution...
what you really get is much higher contrast, so the features look "sharper"
You also get a little bit more resolution.. but thats not really the main point.
Its really about contrast.

> I would expect something similar for the OMX when the reconstruction is pe=
> rfect and includes proper deconvolution. However we get larger features,=
> still within the expected OMX resolution 120 nm in width and 300 nm in z,=
> but no dots.

No reconstruction or deconvolution will give images that contain features that are smaller
than the resolution limit of the system.

What do you mean by "dots" ?
Single 40 nm pixels?

In the OMX it's twice the conventional resolution,
so you cant get objects that appear smaller than 120 nm.
(actually... API say the OMX system sometimes seems to outperform theory slightly... but only a bit)

You should not an must not get single pixel object with the reconstructed pixel spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x resolution increase from the physical 80 nm pixel spacing)

If you get smaller objects, its wrong, and an artifact. No?

> I think there is still some improvement possible either in=
> the setup of the instrument or the software.

Actually, in my hands it seems to do what theory predicts and no more.
There is no good reason to expect it to make smaller objects than 120 nm.

Certainly the careful alignment of the optics and the highest quality objective lens are critical,
as is the measurement of suitable SIM PSFs  and careful calibration/measurement of the parameters for the reconstruction...
the angles and the phases.

So, other than making it faster and more tolerant to sub optimal input data,
I dont think there is any improvement to be made in resolution.. so I dont see how you think there is room for resolution improvement?
Maybe you can share your ideas?

cheers

Dan


>
> best wishes
>
> Andreas
>
>
> =20
>
> =20

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 )

[Northan, Brian]

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

Commercial response from Media Cybernetics:

"The images of sub resolution objects dont get very much smaller after deconvolution...
what you really get is much higher contrast, so the features look "sharper"
You also get a little bit more resolution.. but thats not really the main point.
Its really about contrast."

Daniel.  This is exactly right and a very important point.  It is about contrast and also noise removal.  

"If you get smaller objects, its wrong, and an artifact. No?"

If you got a smaller objects there is no certainty that it is really that size.  In this case the image has been "over-deconvolved".  In these images there can be artifacts.  And very faint, small but real features can be eliminated.

Brian

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Daniel James White
Sent: Monday, April 11, 2011 5:53 AM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

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

Hi Andreas,

On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system wrote:

> Date:    Fri, 8 Apr 2011 11:48:46 -0400
> From:    Andreas Bruckbauer <[hidden email]>
> Subject: Re: Deconvolution of 3D SIM data
>
>
> When i generate a PSF for deconvolution using suitable beads, then image=
> the same beads again and deconvolve the image, i would expect to get real=
> ly tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample=20

real tiny? The images of sub resolution objects dont get very much smaller after deconvolution...
what you really get is much higher contrast, so the features look "sharper"
You also get a little bit more resolution.. but thats not really the main point.
Its really about contrast.

> I would expect something similar for the OMX when the reconstruction is pe=
> rfect and includes proper deconvolution. However we get larger features,=
> still within the expected OMX resolution 120 nm in width and 300 nm in z,=
> but no dots.

No reconstruction or deconvolution will give images that contain features that are smaller
than the resolution limit of the system.

What do you mean by "dots" ?
Single 40 nm pixels?

In the OMX it's twice the conventional resolution,
so you cant get objects that appear smaller than 120 nm.
(actually... API say the OMX system sometimes seems to outperform theory slightly... but only a bit)

You should not an must not get single pixel object with the reconstructed pixel spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x resolution increase from the physical 80 nm pixel spacing)

If you get smaller objects, its wrong, and an artifact. No?

> I think there is still some improvement possible either in=
> the setup of the instrument or the software.

Actually, in my hands it seems to do what theory predicts and no more.
There is no good reason to expect it to make smaller objects than 120 nm.

Certainly the careful alignment of the optics and the highest quality objective lens are critical,
as is the measurement of suitable SIM PSFs  and careful calibration/measurement of the parameters for the reconstruction...
the angles and the phases.

So, other than making it faster and more tolerant to sub optimal input data,
I dont think there is any improvement to be made in resolution.. so I dont see how you think there is room for resolution improvement?
Maybe you can share your ideas?

cheers

Dan


>
> best wishes
>
> Andreas
>
>
> =20
>
> =20

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 )
######################################################################################
CONFIDENTIALITY NOTICE:
This email transmission and its attachments contain confidential and proprietary information
of Princeton Instruments, Acton Research, Media Cybernetics and their affiliates and is
intended for the exclusive and confidential use of the intended recipient. Any use, dissemination,
printing, or copying of this transmission and its attachment(s) is strictly prohibited. If you
are not the intended recipient, please do not read, print, copy, distribute or take action in
reliance upon this message.  If you have received this in error, please notify the sender immediately
by telephone or return email and promptly delete all copies of the original transmission and its
attachments from your computer system.
#######################################################################################

[Andreas Bruckbauer]

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


Hi Dan, Brian

thanks for the comments, i know that for real live samples everything what exceeds the resolution limit of the system is an artefact, however for sparsely separated beads i would expect that i can determine their position with high accuracy only limited by the signal to noise ratio. This is exactly what is done in localisation microscopy (PALM, STORM...). One approach is to calculate the cross correlation between the measured 3D PSF and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011, 327). While this is not exactly what deconvolution does, it is at least related to deconvolution.

So i was wondering if this special case, a limited number of well resolved beads which are small enough to qualify as point objects, could tell us how good the deconvolution method actually works, or in the case of SIM, what the image restauration actually does. At least theoretically, the size of the beads (110 nm in my case) should not be a problem as by using the same beads to measure the PSF we trick the deconvolution into thinking that they are point objects. This is similar to the obove mentioned localisation microscopy which does not depend much on optical resolution or pixel size.

best wishes

Andreas



 

 


 

 

-----Original Message-----
From: Daniel James White <[hidden email]>
To: [hidden email]
Sent: Mon, 11 Apr 2011 10:53
Subject: Re: Deconvolution of 3D SIM data


*****





To join, leave or search the confocal microscopy listserv, go to:





http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy





*****











Hi Andreas,











On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system wrote:











> Date:    Fri, 8 Apr 2011 11:48:46 -0400





> From:    Andreas Bruckbauer <[hidden email]>





> Subject: Re: Deconvolution of 3D SIM data





>





>





> When i generate a PSF for deconvolution using suitable beads, then image=





> the same beads again and deconvolve the image, i would expect to get real=





> ly tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample=20











real tiny? The images of sub resolution objects dont get very much smaller after





deconvolution...





what you really get is much higher contrast, so the features look "sharper"





You also get a little bit more resolution.. but thats not really the main point.











Its really about contrast.











> I would expect something similar for the OMX when the reconstruction is pe=





> rfect and includes proper deconvolution. However we get larger features,=





> still within the expected OMX resolution 120 nm in width and 300 nm in z,=





> but no dots.











No reconstruction or deconvolution will give images that contain features that





are smaller





than the resolution limit of the system.











What do you mean by "dots" ?





Single 40 nm pixels?











In the OMX it's twice the conventional resolution,





so you cant get objects that appear smaller than 120 nm.





(actually... API say the OMX system sometimes seems to outperform theory





slightly... but only a bit)











You should not an must not get single pixel object with the reconstructed pixel





spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x resolution





increase from the physical 80 nm pixel spacing)











If you get smaller objects, its wrong, and an artifact. No?











> I think there is still some improvement possible either in=





> the setup of the instrument or the software.











Actually, in my hands it seems to do what theory predicts and no more.





There is no good reason to expect it to make smaller objects than 120 nm.











Certainly the careful alignment of the optics and the highest quality objective





lens are critical,





as is the measurement of suitable SIM PSFs  and careful calibration/measurement





of the parameters for the reconstruction...





the angles and the phases.











So, other than making it faster and more tolerant to sub optimal input data,





I dont think there is any improvement to be made in resolution.. so I dont see





how you think there is room for resolution improvement?





Maybe you can share your ideas?











cheers











Dan

















>





> best wishes





>





> Andreas





>





>





> =20





>





> =20











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 )






 
 

[lechristophe]

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

To elaborate a bit on that, why isn't it possible to go beyond the
diffraction limit with deconvolution ? I don't see the theoretical reason,
in a ideal case (subresolutive beads sufficiently spaced). Of course in a
real sample, S/N issues and fluorophore density impose a limit on the
precision, but why would this limit precisely be at the diffraction limit ?

Christophe

On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
>
> Hi Dan, Brian
>
> thanks for the comments, i know that for real live samples everything what
> exceeds the resolution limit of the system is an artefact, however for
> sparsely separated beads i would expect that i can determine their position
> with high accuracy only limited by the signal to noise ratio. This is
> exactly what is done in localisation microscopy (PALM, STORM...). One
> approach is to calculate the cross correlation between the measured 3D PSF
> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011, 327).
> While this is not exactly what deconvolution does, it is at least related to
> deconvolution.
>
> So i was wondering if this special case, a limited number of well resolved
> beads which are small enough to qualify as point objects, could tell us how
> good the deconvolution method actually works, or in the case of SIM, what
> the image restauration actually does. At least theoretically, the size of
> the beads (110 nm in my case) should not be a problem as by using the same
> beads to measure the PSF we trick the deconvolution into thinking that they
> are point objects. This is similar to the obove mentioned localisation
> microscopy which does not depend much on optical resolution or pixel size.
>
> best wishes
>
> Andreas
>
>
>
>
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: Daniel James White <[hidden email]>
> To: [hidden email]
> Sent: Mon, 11 Apr 2011 10:53
> Subject: Re: Deconvolution of 3D SIM data
>
>
> *****
>
>
>
>
>
> To join, leave or search the confocal microscopy listserv, go to:
>
>
>
>
>
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>
>
>
>
>
> *****
>
>
>
>
>
>
>
>
>
>
>
> Hi Andreas,
>
>
>
>
>
>
>
>
>
>
>
> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
> wrote:
>
>
>
>
>
>
>
>
>
>
>
> > Date:    Fri, 8 Apr 2011 11:48:46 -0400
>
>
>
>
>
> > From:    Andreas Bruckbauer <[hidden email]>
>
>
>
>
>
> > Subject: Re: Deconvolution of 3D SIM data
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > When i generate a PSF for deconvolution using suitable beads, then image=
>
>
>
>
>
> > the same beads again and deconvolve the image, i would expect to get
> real=
>
>
>
>
>
> > ly tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample=20
>
>
>
>
>
>
>
>
>
>
>
> real tiny? The images of sub resolution objects dont get very much smaller
> after
>
>
>
>
>
> deconvolution...
>
>
>
>
>
> what you really get is much higher contrast, so the features look "sharper"
>
>
>
>
>
> You also get a little bit more resolution.. but thats not really the main
> point.
>
>
>
>
>
>
>
>
>
>
>
> Its really about contrast.
>
>
>
>
>
>
>
>
>
>
>
> > I would expect something similar for the OMX when the reconstruction is
> pe=
>
>
>
>
>
> > rfect and includes proper deconvolution. However we get larger features,=
>
>
>
>
>
> > still within the expected OMX resolution 120 nm in width and 300 nm in
> z,=
>
>
>
>
>
> > but no dots.
>
>
>
>
>
>
>
>
>
>
>
> No reconstruction or deconvolution will give images that contain features
> that
>
>
>
>
>
> are smaller
>
>
>
>
>
> than the resolution limit of the system.
>
>
>
>
>
>
>
>
>
>
>
> What do you mean by "dots" ?
>
>
>
>
>
> Single 40 nm pixels?
>
>
>
>
>
>
>
>
>
>
>
> In the OMX it's twice the conventional resolution,
>
>
>
>
>
> so you cant get objects that appear smaller than 120 nm.
>
>
>
>
>
> (actually... API say the OMX system sometimes seems to outperform theory
>
>
>
>
>
> slightly... but only a bit)
>
>
>
>
>
>
>
>
>
>
>
> You should not an must not get single pixel object with the reconstructed
> pixel
>
>
>
>
>
> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
> resolution
>
>
>
>
>
> increase from the physical 80 nm pixel spacing)
>
>
>
>
>
>
>
>
>
>
>
> If you get smaller objects, its wrong, and an artifact. No?
>
>
>
>
>
>
>
>
>
>
>
> > I think there is still some improvement possible either in=
>
>
>
>
>
> > the setup of the instrument or the software.
>
>
>
>
>
>
>
>
>
>
>
> Actually, in my hands it seems to do what theory predicts and no more.
>
>
>
>
>
> There is no good reason to expect it to make smaller objects than 120 nm.
>
>
>
>
>
>
>
>
>
>
>
> Certainly the careful alignment of the optics and the highest quality
> objective
>
>
>
>
>
> lens are critical,
>
>
>
>
>
> as is the measurement of suitable SIM PSFs  and careful
> calibration/measurement
>
>
>
>
>
> of the parameters for the reconstruction...
>
>
>
>
>
> the angles and the phases.
>
>
>
>
>
>
>
>
>
>
>
> So, other than making it faster and more tolerant to sub optimal input
> data,
>
>
>
>
>
> I dont think there is any improvement to be made in resolution.. so I dont
> see
>
>
>
>
>
> how you think there is room for resolution improvement?
>
>
>
>
>
> Maybe you can share your ideas?
>
>
>
>
>
>
>
>
>
>
>
> cheers
>
>
>
>
>
>
>
>
>
>
>
> Dan
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> >
>
>
>
>
>
> > best wishes
>
>
>
>
>
> >
>
>
>
>
>
> > Andreas
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
>
>
>
>
>
>
> 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 )
>
>
>
>
>
>
>
>
>

[David Baddeley]

the diffraction limit means that your raw data contains
*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi Christophe,

the diffraction limit means that your raw data contains NO information for
spatial frequencies above the limit. Regardless of the the SNR or sampling you
have zero power in spatial frequencies above the diffraction limit.
Deconvolution algorithms can only make an informed guess at what this out of
band information should be (typically based on positivity and smoothness
constraints), and whilst that is usually sufficient to recover enough
information over the diffraction limit to prevent the high frequency ripple you
would get if you only restored just the 'in band' frequencies, it won't give you
reliable information about structures in the regime.

cheers,
David


----- Original Message ----
From: Christophe Leterrier <[hidden email]>
To: [hidden email]
Sent: Wed, 13 April, 2011 9:31:18 PM
Subject: Re: Deconvolution of 3D SIM data

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

To elaborate a bit on that, why isn't it possible to go beyond the
diffraction limit with deconvolution ? I don't see the theoretical reason,
in a ideal case (subresolutive beads sufficiently spaced). Of course in a
real sample, S/N issues and fluorophore density impose a limit on the
precision, but why would this limit precisely be at the diffraction limit ?

Christophe

On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
>
> Hi Dan, Brian
>
> thanks for the comments, i know that for real live samples everything what
> exceeds the resolution limit of the system is an artefact, however for
> sparsely separated beads i would expect that i can determine their position
> with high accuracy only limited by the signal to noise ratio. This is
> exactly what is done in localisation microscopy (PALM, STORM...). One
> approach is to calculate the cross correlation between the measured 3D PSF
> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011, 327).
> While this is not exactly what deconvolution does, it is at least related to
> deconvolution.
>
> So i was wondering if this special case, a limited number of well resolved
> beads which are small enough to qualify as point objects, could tell us how
> good the deconvolution method actually works, or in the case of SIM, what
> the image restauration actually does. At least theoretically, the size of
> the beads (110 nm in my case) should not be a problem as by using the same
> beads to measure the PSF we trick the deconvolution into thinking that they
> are point objects. This is similar to the obove mentioned localisation
> microscopy which does not depend much on optical resolution or pixel size.
>
> best wishes
>
> Andreas
>
>
>
>
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: Daniel James White <[hidden email]>
> To: [hidden email]
> Sent: Mon, 11 Apr 2011 10:53
> Subject: Re: Deconvolution of 3D SIM data
>
>
> *****
>
>
>
>
>
> To join, leave or search the confocal microscopy listserv, go to:
>
>
>
>
>
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>
>
>
>
>
> *****
>
>
>
>
>
>
>
>
>
>
>
> Hi Andreas,
>
>
>
>
>
>
>
>
>
>
>
> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
> wrote:
>
>
>
>
>
>
>
>
>
>
>
> > Date:    Fri, 8 Apr 2011 11:48:46 -0400
>
>
>
>
>
> > From:    Andreas Bruckbauer <[hidden email]>
>
>
>
>
>
> > Subject: Re: Deconvolution of 3D SIM data
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > When i generate a PSF for deconvolution using suitable beads, then image=
>
>
>
>
>
> > the same beads again and deconvolve the image, i would expect to get
> real=
>
>
>
>
>
> > ly tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample=20
>
>
>
>
>
>
>
>
>
>
>
> real tiny? The images of sub resolution objects dont get very much smaller
> after
>
>
>
>
>
> deconvolution...
>
>
>
>
>
> what you really get is much higher contrast, so the features look "sharper"
>
>
>
>
>
> You also get a little bit more resolution.. but thats not really the main
> point.
>
>
>
>
>
>
>
>
>
>
>
> Its really about contrast.
>
>
>
>
>
>
>
>
>
>
>
> > I would expect something similar for the OMX when the reconstruction is
> pe=
>
>
>
>
>
> > rfect and includes proper deconvolution. However we get larger features,=
>
>
>
>
>
> > still within the expected OMX resolution 120 nm in width and 300 nm in
> z,=
>
>
>
>
>
> > but no dots.
>
>
>
>
>
>
>
>
>
>
>
> No reconstruction or deconvolution will give images that contain features
> that
>
>
>
>
>
> are smaller
>
>
>
>
>
> than the resolution limit of the system.
>
>
>
>
>
>
>
>
>
>
>
> What do you mean by "dots" ?
>
>
>
>
>
> Single 40 nm pixels?
>
>
>
>
>
>
>
>
>
>
>
> In the OMX it's twice the conventional resolution,
>
>
>
>
>
> so you cant get objects that appear smaller than 120 nm.
>
>
>
>
>
> (actually... API say the OMX system sometimes seems to outperform theory
>
>
>
>
>
> slightly... but only a bit)
>
>
>
>
>
>
>
>
>
>
>
> You should not an must not get single pixel object with the reconstructed
> pixel
>
>
>
>
>
> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
> resolution
>
>
>
>
>
> increase from the physical 80 nm pixel spacing)
>
>
>
>
>
>
>
>
>
>
>
> If you get smaller objects, its wrong, and an artifact. No?
>
>
>
>
>
>
>
>
>
>
>
> > I think there is still some improvement possible either in=
>
>
>
>
>
> > the setup of the instrument or the software.
>
>
>
>
>
>
>
>
>
>
>
> Actually, in my hands it seems to do what theory predicts and no more.
>
>
>
>
>
> There is no good reason to expect it to make smaller objects than 120 nm.
>
>
>
>
>
>
>
>
>
>
>
> Certainly the careful alignment of the optics and the highest quality
> objective
>
>
>
>
>
> lens are critical,
>
>
>
>
>
> as is the measurement of suitable SIM PSFs  and careful
> calibration/measurement
>
>
>
>
>
> of the parameters for the reconstruction...
>
>
>
>
>
> the angles and the phases.
>
>
>
>
>
>
>
>
>
>
>
> So, other than making it faster and more tolerant to sub optimal input
> data,
>
>
>
>
>
> I dont think there is any improvement to be made in resolution.. so I dont
> see
>
>
>
>
>
> how you think there is room for resolution improvement?
>
>
>
>
>
> Maybe you can share your ideas?
>
>
>
>
>
>
>
>
>
>
>
> cheers
>
>
>
>
>
>
>
>
>
>
>
> Dan
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> >
>
>
>
>
>
> > best wishes
>
>
>
>
>
> >
>
>
>
>
>
> > Andreas
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
>
>
>
>
>
>
> 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 )
>
>
>
>
>
>
>
>
>

[Nuno Moreno]

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

Hi,

I'm not going to give an answer, rather to raise some theoretical aspect of deconvolution and SIM and give my non scientific opinion.

As far as I know, in noise free images, making deconvolution would be as easy as making a division of the image with the PSF in the frequency domain, as convolution in frequency domain is a simple multiplication of the real image with the characterization of the objective...PSF. Unfortunately, due to noise, this is not straightforward but the goal should be the same.

In SIM, images from the frequency domain containing the Moiré patterns are put back together using the interference of typical 9 different angles/phase shifts, from which high frequency information is retrieved. Therefore, the "behavior" of each dot as a PSF for classical deconvolution makes no sense to me (eventually with 9 PSFs...and deconvolve each before reconstruction...). Furthermore, SIM only works for features which are in the interference range, i.e., resolution limit, boosting on those cases the MTF (Modulation transfer function), having now a new local maximum and change the supposed linearity of deconv. Also, as it was already posted here, it is possible to use non linear Fourier reconstruction to go a little further.

Therefore, I would be very, very skeptical either on any deconvolution and quantification using SIM.


Nuno Moreno, PhD
Head of Equipment Management
Instituto Gulbenkian de Ciência
Fundação Calouste Gulbenkian
Tel . [phone]
Fax. [phone]





On Apr 13, 2011, at 10:31 AM, Christophe Leterrier wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> To elaborate a bit on that, why isn't it possible to go beyond the
> diffraction limit with deconvolution ? I don't see the theoretical reason,
> in a ideal case (subresolutive beads sufficiently spaced). Of course in a
> real sample, S/N issues and fluorophore density impose a limit on the
> precision, but why would this limit precisely be at the diffraction limit ?
>
> Christophe
>
> On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]> wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>>
>>
>> Hi Dan, Brian
>>
>> thanks for the comments, i know that for real live samples everything what
>> exceeds the resolution limit of the system is an artefact, however for
>> sparsely separated beads i would expect that i can determine their position
>> with high accuracy only limited by the signal to noise ratio. This is
>> exactly what is done in localisation microscopy (PALM, STORM...). One
>> approach is to calculate the cross correlation between the measured 3D PSF
>> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011, 327).
>> While this is not exactly what deconvolution does, it is at least related to
>> deconvolution.
>>
>> So i was wondering if this special case, a limited number of well resolved
>> beads which are small enough to qualify as point objects, could tell us how
>> good the deconvolution method actually works, or in the case of SIM, what
>> the image restauration actually does. At least theoretically, the size of
>> the beads (110 nm in my case) should not be a problem as by using the same
>> beads to measure the PSF we trick the deconvolution into thinking that they
>> are point objects. This is similar to the obove mentioned localisation
>> microscopy which does not depend much on optical resolution or pixel size.
>>
>> best wishes
>>
>> Andreas
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> -----Original Message-----
>> From: Daniel James White <[hidden email]>
>> To: [hidden email]
>> Sent: Mon, 11 Apr 2011 10:53
>> Subject: Re: Deconvolution of 3D SIM data
>>
>>
>> *****
>>
>>
>>
>>
>>
>> To join, leave or search the confocal microscopy listserv, go to:
>>
>>
>>
>>
>>
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>
>>
>>
>>
>>
>> *****
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Hi Andreas,
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
>> wrote:
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>> Date:    Fri, 8 Apr 2011 11:48:46 -0400
>>
>>
>>
>>
>>
>>> From:    Andreas Bruckbauer <[hidden email]>
>>
>>
>>
>>
>>
>>> Subject: Re: Deconvolution of 3D SIM data
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> When i generate a PSF for deconvolution using suitable beads, then image=
>>
>>
>>
>>
>>
>>> the same beads again and deconvolve the image, i would expect to get
>> real=
>>
>>
>>
>>
>>
>>> ly tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample=20
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> real tiny? The images of sub resolution objects dont get very much smaller
>> after
>>
>>
>>
>>
>>
>> deconvolution...
>>
>>
>>
>>
>>
>> what you really get is much higher contrast, so the features look "sharper"
>>
>>
>>
>>
>>
>> You also get a little bit more resolution.. but thats not really the main
>> point.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Its really about contrast.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>> I would expect something similar for the OMX when the reconstruction is
>> pe=
>>
>>
>>
>>
>>
>>> rfect and includes proper deconvolution. However we get larger features,=
>>
>>
>>
>>
>>
>>> still within the expected OMX resolution 120 nm in width and 300 nm in
>> z,=
>>
>>
>>
>>
>>
>>> but no dots.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> No reconstruction or deconvolution will give images that contain features
>> that
>>
>>
>>
>>
>>
>> are smaller
>>
>>
>>
>>
>>
>> than the resolution limit of the system.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> What do you mean by "dots" ?
>>
>>
>>
>>
>>
>> Single 40 nm pixels?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> In the OMX it's twice the conventional resolution,
>>
>>
>>
>>
>>
>> so you cant get objects that appear smaller than 120 nm.
>>
>>
>>
>>
>>
>> (actually... API say the OMX system sometimes seems to outperform theory
>>
>>
>>
>>
>>
>> slightly... but only a bit)
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> You should not an must not get single pixel object with the reconstructed
>> pixel
>>
>>
>>
>>
>>
>> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
>> resolution
>>
>>
>>
>>
>>
>> increase from the physical 80 nm pixel spacing)
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> If you get smaller objects, its wrong, and an artifact. No?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>> I think there is still some improvement possible either in=
>>
>>
>>
>>
>>
>>> the setup of the instrument or the software.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Actually, in my hands it seems to do what theory predicts and no more.
>>
>>
>>
>>
>>
>> There is no good reason to expect it to make smaller objects than 120 nm.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Certainly the careful alignment of the optics and the highest quality
>> objective
>>
>>
>>
>>
>>
>> lens are critical,
>>
>>
>>
>>
>>
>> as is the measurement of suitable SIM PSFs  and careful
>> calibration/measurement
>>
>>
>>
>>
>>
>> of the parameters for the reconstruction...
>>
>>
>>
>>
>>
>> the angles and the phases.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> So, other than making it faster and more tolerant to sub optimal input
>> data,
>>
>>
>>
>>
>>
>> I dont think there is any improvement to be made in resolution.. so I dont
>> see
>>
>>
>>
>>
>>
>> how you think there is room for resolution improvement?
>>
>>
>>
>>
>>
>> Maybe you can share your ideas?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> cheers
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Dan
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> best wishes
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> Andreas
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> =20
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> =20
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> 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 )
>>
>>
>>
>>
>>
>>
>>
>>
>>

[Guy Cox-2]

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

Hear hear!

 

                       Guy

 

Sponsor my next half-marathon on May 15th

There's a special reason - find it out at

http://www.everydayhero.com.au/Guy_Cox_4846 <http://www.everydayhero.com.au/Guy_Cox_4846>

______________________________________________

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 Nuno Moreno
Sent: Wednesday, 13 April 2011 8:33 PM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

 

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

Hi,

I'm not going to give an answer, rather to raise some theoretical aspect of deconvolution and SIM and give my non scientific opinion.

As far as I know, in noise free images, making deconvolution would be as easy as making a division of the image with the PSF in the frequency domain, as convolution in frequency domain is a simple multiplication of the real image with the characterization of the objective...PSF. Unfortunately, due to noise, this is not straightforward but the goal should be the same.

In SIM, images from the frequency domain containing the Moiré patterns are put back together using the interference of typical 9 different angles/phase shifts, from which high frequency information is retrieved. Therefore, the "behavior" of each dot as a PSF for classical deconvolution makes no sense to me (eventually with 9 PSFs...and deconvolve each before reconstruction...). Furthermore, SIM only works for features which are in the interference range, i.e., resolution limit, boosting on those cases the MTF (Modulation transfer function), having now a new local maximum and change the supposed linearity of deconv. Also, as it was already posted here, it is possible to use non linear Fourier reconstruction to go a little further.

Therefore, I would be very, very skeptical either on any deconvolution and quantification using SIM.


Nuno Moreno, PhD
Head of Equipment Management
Instituto Gulbenkian de Ciência
Fundação Calouste Gulbenkian
Tel . [phone]
Fax. [phone]





On Apr 13, 2011, at 10:31 AM, Christophe Leterrier wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> To elaborate a bit on that, why isn't it possible to go beyond the
> diffraction limit with deconvolution ? I don't see the theoretical reason,
> in a ideal case (subresolutive beads sufficiently spaced). Of course in a
> real sample, S/N issues and fluorophore density impose a limit on the
> precision, but why would this limit precisely be at the diffraction limit ?
>
> Christophe
>
> On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]> wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>>
>>
>> Hi Dan, Brian
>>
>> thanks for the comments, i know that for real live samples everything what
>> exceeds the resolution limit of the system is an artefact, however for
>> sparsely separated beads i would expect that i can determine their position
>> with high accuracy only limited by the signal to noise ratio. This is
>> exactly what is done in localisation microscopy (PALM, STORM...). One
>> approach is to calculate the cross correlation between the measured 3D PSF
>> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011, 327).
>> While this is not exactly what deconvolution does, it is at least related to
>> deconvolution.
>>
>> So i was wondering if this special case, a limited number of well resolved
>> beads which are small enough to qualify as point objects, could tell us how
>> good the deconvolution method actually works, or in the case of SIM, what
>> the image restauration actually does. At least theoretically, the size of
>> the beads (110 nm in my case) should not be a problem as by using the same
>> beads to measure the PSF we trick the deconvolution into thinking that they
>> are point objects. This is similar to the obove mentioned localisation
>> microscopy which does not depend much on optical resolution or pixel size.
>>
>> best wishes
>>
>> Andreas
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> -----Original Message-----
>> From: Daniel James White <[hidden email]>
>> To: [hidden email]
>> Sent: Mon, 11 Apr 2011 10:53
>> Subject: Re: Deconvolution of 3D SIM data
>>
>>
>> *****
>>
>>
>>
>>
>>
>> To join, leave or search the confocal microscopy listserv, go to:
>>
>>
>>
>>
>>
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>
>>
>>
>>
>>
>> *****
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Hi Andreas,
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
>> wrote:
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>> Date:    Fri, 8 Apr 2011 11:48:46 -0400
>>
>>
>>
>>
>>
>>> From:    Andreas Bruckbauer <[hidden email]>
>>
>>
>>
>>
>>
>>> Subject: Re: Deconvolution of 3D SIM data
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> When i generate a PSF for deconvolution using suitable beads, then image=
>>
>>
>>
>>
>>
>>> the same beads again and deconvolve the image, i would expect to get
>> real=
>>
>>
>>
>>
>>
>>> ly tiny dots. See e.g. http://www.svi.nl/BeadsDeconvolutionExample=20
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> real tiny? The images of sub resolution objects dont get very much smaller
>> after
>>
>>
>>
>>
>>
>> deconvolution...
>>
>>
>>
>>
>>
>> what you really get is much higher contrast, so the features look "sharper"
>>
>>
>>
>>
>>
>> You also get a little bit more resolution.. but thats not really the main
>> point.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Its really about contrast.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>> I would expect something similar for the OMX when the reconstruction is
>> pe=
>>
>>
>>
>>
>>
>>> rfect and includes proper deconvolution. However we get larger features,=
>>
>>
>>
>>
>>
>>> still within the expected OMX resolution 120 nm in width and 300 nm in
>> z,=
>>
>>
>>
>>
>>
>>> but no dots.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> No reconstruction or deconvolution will give images that contain features
>> that
>>
>>
>>
>>
>>
>> are smaller
>>
>>
>>
>>
>>
>> than the resolution limit of the system.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> What do you mean by "dots" ?
>>
>>
>>
>>
>>
>> Single 40 nm pixels?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> In the OMX it's twice the conventional resolution,
>>
>>
>>
>>
>>
>> so you cant get objects that appear smaller than 120 nm.
>>
>>
>>
>>
>>
>> (actually... API say the OMX system sometimes seems to outperform theory
>>
>>
>>
>>
>>
>> slightly... but only a bit)
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> You should not an must not get single pixel object with the reconstructed
>> pixel
>>
>>
>>
>>
>>
>> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
>> resolution
>>
>>
>>
>>
>>
>> increase from the physical 80 nm pixel spacing)
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> If you get smaller objects, its wrong, and an artifact. No?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>> I think there is still some improvement possible either in=
>>
>>
>>
>>
>>
>>> the setup of the instrument or the software.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Actually, in my hands it seems to do what theory predicts and no more.
>>
>>
>>
>>
>>
>> There is no good reason to expect it to make smaller objects than 120 nm.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Certainly the careful alignment of the optics and the highest quality
>> objective
>>
>>
>>
>>
>>
>> lens are critical,
>>
>>
>>
>>
>>
>> as is the measurement of suitable SIM PSFs  and careful
>> calibration/measurement
>>
>>
>>
>>
>>
>> of the parameters for the reconstruction...
>>
>>
>>
>>
>>
>> the angles and the phases.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> So, other than making it faster and more tolerant to sub optimal input
>> data,
>>
>>
>>
>>
>>
>> I dont think there is any improvement to be made in resolution.. so I dont
>> see
>>
>>
>>
>>
>>
>> how you think there is room for resolution improvement?
>>
>>
>>
>>
>>
>> Maybe you can share your ideas?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> cheers
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Dan
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> best wishes
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> Andreas
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> =20
>>
>>
>>
>>
>>
>>>
>>
>>
>>
>>
>>
>>> =20
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> 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 )
>>
>>
>>
>>
>>
>>
>>
>>
>>

________________________________

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11

[Guy Cox-2]

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

Well, the critical point is the 'sufficiently spaced' bit.  If your
points are well separated you can find their centroids with a precision
way beyond the Rayleigh diffraction limit, which essentially applies to
objects separated by the diffraction limit.  Rayleigh's criterion is
only a rule of thumb - with sufficient signal to noise you can separate
objects closer together than the limit by deconvolution.   Abbe's limit,
on the other hand, is non-negotiable, but it doesn't apply in
fluorescence.  However, when it comes to SIM data, as Nuno has pointed
out, we have already defined the bandwith we are probing, and there is
nothing more to get.

 

 
Guy

 

Sponsor my next half-marathon on May 15th

There's a special reason - find it out at

http://www.everydayhero.com.au/Guy_Cox_4846
<http://www.everydayhero.com.au/Guy_Cox_4846>

______________________________________________

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 Christophe Leterrier
Sent: Wednesday, 13 April 2011 7:31 PM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

 

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

To elaborate a bit on that, why isn't it possible to go beyond the
diffraction limit with deconvolution ? I don't see the theoretical
reason,
in a ideal case (subresolutive beads sufficiently spaced). Of course in
a
real sample, S/N issues and fluorophore density impose a limit on the
precision, but why would this limit precisely be at the diffraction
limit ?

Christophe

On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]>
wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
>
> Hi Dan, Brian
>
> thanks for the comments, i know that for real live samples everything
what
> exceeds the resolution limit of the system is an artefact, however for
> sparsely separated beads i would expect that i can determine their
position
> with high accuracy only limited by the signal to noise ratio. This is
> exactly what is done in localisation microscopy (PALM, STORM...). One
> approach is to calculate the cross correlation between the measured 3D
PSF
> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011,
327).
> While this is not exactly what deconvolution does, it is at least
related to
> deconvolution.
>
> So i was wondering if this special case, a limited number of well
resolved
> beads which are small enough to qualify as point objects, could tell
us how
> good the deconvolution method actually works, or in the case of SIM,
what
> the image restauration actually does. At least theoretically, the size
of
> the beads (110 nm in my case) should not be a problem as by using the
same
> beads to measure the PSF we trick the deconvolution into thinking that
they
> are point objects. This is similar to the obove mentioned localisation
> microscopy which does not depend much on optical resolution or pixel
size.

>
> best wishes
>
> Andreas
>
>
>
>
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: Daniel James White <[hidden email]>
> To: [hidden email]
> Sent: Mon, 11 Apr 2011 10:53
> Subject: Re: Deconvolution of 3D SIM data
>
>
> *****
>
>
>
>
>
> To join, leave or search the confocal microscopy listserv, go to:
>
>
>
>
>
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>
>
>
>
>
> *****
>
>
>
>
>
>
>
>
>
>
>
> Hi Andreas,
>
>
>
>
>
>
>
>
>
>
>
> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
> wrote:
>
>
>
>
>
>
>
>
>
>
>
> > Date:    Fri, 8 Apr 2011 11:48:46 -0400
>
>
>
>
>
> > From:    Andreas Bruckbauer <[hidden email]>
>
>
>
>
>
> > Subject: Re: Deconvolution of 3D SIM data
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > When i generate a PSF for deconvolution using suitable beads, then

image=

>
>
>
>
>
> > the same beads again and deconvolve the image, i would expect to get
> real=
>
>
>
>
>
> > ly tiny dots. See e.g.

http://www.svi.nl/BeadsDeconvolutionExample=20

>
>
>
>
>
>
>
>
>
>
>
> real tiny? The images of sub resolution objects dont get very much

smaller

> after
>
>
>
>
>
> deconvolution...
>
>
>
>
>
> what you really get is much higher contrast, so the features look

"sharper"
>
>
>
>
>
> You also get a little bit more resolution.. but thats not really the
main

> point.
>
>
>
>
>
>
>
>
>
>
>
> Its really about contrast.
>
>
>
>
>
>
>
>
>
>
>
> > I would expect something similar for the OMX when the reconstruction

is
> pe=
>
>
>
>
>
> > rfect and includes proper deconvolution. However we get larger
features,=
>
>
>
>
>
> > still within the expected OMX resolution 120 nm in width and 300 nm
in

> z,=
>
>
>
>
>
> > but no dots.
>
>
>
>
>
>
>
>
>
>
>
> No reconstruction or deconvolution will give images that contain

features

> that
>
>
>
>
>
> are smaller
>
>
>
>
>
> than the resolution limit of the system.
>
>
>
>
>
>
>
>
>
>
>
> What do you mean by "dots" ?
>
>
>
>
>
> Single 40 nm pixels?
>
>
>
>
>
>
>
>
>
>
>
> In the OMX it's twice the conventional resolution,
>
>
>
>
>
> so you cant get objects that appear smaller than 120 nm.
>
>
>
>
>
> (actually... API say the OMX system sometimes seems to outperform

theory

>
>
>
>
>
> slightly... but only a bit)
>
>
>
>
>
>
>
>
>
>
>
> You should not an must not get single pixel object with the

reconstructed

> pixel
>
>
>
>
>
> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
> resolution
>
>
>
>
>
> increase from the physical 80 nm pixel spacing)
>
>
>
>
>
>
>
>
>
>
>
> If you get smaller objects, its wrong, and an artifact. No?
>
>
>
>
>
>
>
>
>
>
>
> > I think there is still some improvement possible either in=
>
>
>
>
>
> > the setup of the instrument or the software.
>
>
>
>
>
>
>
>
>
>
>
> Actually, in my hands it seems to do what theory predicts and no more.
>
>
>
>
>
> There is no good reason to expect it to make smaller objects than 120

nm.

>
>
>
>
>
>
>
>
>
>
>
> Certainly the careful alignment of the optics and the highest quality
> objective
>
>
>
>
>
> lens are critical,
>
>
>
>
>
> as is the measurement of suitable SIM PSFs  and careful
> calibration/measurement
>
>
>
>
>
> of the parameters for the reconstruction...
>
>
>
>
>
> the angles and the phases.
>
>
>
>
>
>
>
>
>
>
>
> So, other than making it faster and more tolerant to sub optimal input
> data,
>
>
>
>
>
> I dont think there is any improvement to be made in resolution.. so I

dont

> see
>
>
>
>
>
> how you think there is room for resolution improvement?
>
>
>
>
>
> Maybe you can share your ideas?
>
>
>
>
>
>
>
>
>
>
>
> cheers
>
>
>
>
>
>
>
>
>
>
>
> Dan
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> >
>
>
>
>
>
> > best wishes
>
>
>
>
>
> >
>
>
>
>
>
> > Andreas
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
>
>
>
>
>
>
> 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 )
>
>
>
>
>
>
>
>
>

________________________________

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11

[Martin Wessendorf-2]

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

On 4/13/2011 7:33 AM, Guy Cox wrote:

> Abbe's limit,
> on the other hand, is non-negotiable, but it doesn't apply in
> fluorescence.

Guy--can you elaborate a bit on your statement that, "it doesn't apply
in fluorescence"?  Are you talking about STED, or talking in more
general terms?

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]

[Emmanuel Gustin]

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

I think that actually, the critical point is that you *know* the beads
(or in the PALM/STORM case, molecules) are sub-resolution in size? The
image reconstruction uses more information than is present in the image,
and this allows you to go below the optical resolution limit. But there
is no way to tell from the image alone what size the beads are.

I guess that theoretically, if you would feed an algorithm with the
knowledge that you have labeled, for example, tubulin and information
about the patterns you expect to see (size of structures, persistence
lengths of bending, branching behavior, ...) you could use it to do
a type of "deconvolution plus" and create a sub-resolution image.

Whether other people would have confidence in the image that has been
constructed in that way, is another matter.

Best Regards,

Emmanuel


--
 Emmanuel Gustin,    Tel. [phone],    e-mail: [hidden email]  


-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Guy Cox
Sent: woensdag 13 april 2011 14:34
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

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

Well, the critical point is the 'sufficiently spaced' bit.  If your
points are well separated you can find their centroids with a precision
way beyond the Rayleigh diffraction limit, which essentially applies to
objects separated by the diffraction limit.  Rayleigh's criterion is
only a rule of thumb - with sufficient signal to noise you can separate
objects closer together than the limit by deconvolution.   Abbe's limit,
on the other hand, is non-negotiable, but it doesn't apply in
fluorescence.  However, when it comes to SIM data, as Nuno has pointed
out, we have already defined the bandwith we are probing, and there is
nothing more to get.

 

 
Guy

 

Sponsor my next half-marathon on May 15th

There's a special reason - find it out at

http://www.everydayhero.com.au/Guy_Cox_4846
<http://www.everydayhero.com.au/Guy_Cox_4846>

______________________________________________

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 Christophe Leterrier
Sent: Wednesday, 13 April 2011 7:31 PM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

 

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

To elaborate a bit on that, why isn't it possible to go beyond the
diffraction limit with deconvolution ? I don't see the theoretical
reason,
in a ideal case (subresolutive beads sufficiently spaced). Of course in
a
real sample, S/N issues and fluorophore density impose a limit on the
precision, but why would this limit precisely be at the diffraction
limit ?

Christophe

On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]>
wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
>
> Hi Dan, Brian
>
> thanks for the comments, i know that for real live samples everything
what
> exceeds the resolution limit of the system is an artefact, however for
> sparsely separated beads i would expect that i can determine their
position
> with high accuracy only limited by the signal to noise ratio. This is
> exactly what is done in localisation microscopy (PALM, STORM...). One
> approach is to calculate the cross correlation between the measured 3D
PSF
> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011,
327).
> While this is not exactly what deconvolution does, it is at least
related to
> deconvolution.
>
> So i was wondering if this special case, a limited number of well
resolved
> beads which are small enough to qualify as point objects, could tell
us how
> good the deconvolution method actually works, or in the case of SIM,
what
> the image restauration actually does. At least theoretically, the size
of
> the beads (110 nm in my case) should not be a problem as by using the
same
> beads to measure the PSF we trick the deconvolution into thinking that
they
> are point objects. This is similar to the obove mentioned localisation
> microscopy which does not depend much on optical resolution or pixel
size.

>
> best wishes
>
> Andreas
>
>
>
>
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: Daniel James White <[hidden email]>
> To: [hidden email]
> Sent: Mon, 11 Apr 2011 10:53
> Subject: Re: Deconvolution of 3D SIM data
>
>
> *****
>
>
>
>
>
> To join, leave or search the confocal microscopy listserv, go to:
>
>
>
>
>
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>
>
>
>
>
> *****
>
>
>
>
>
>
>
>
>
>
>
> Hi Andreas,
>
>
>
>
>
>
>
>
>
>
>
> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
> wrote:
>
>
>
>
>
>
>
>
>
>
>
> > Date:    Fri, 8 Apr 2011 11:48:46 -0400
>
>
>
>
>
> > From:    Andreas Bruckbauer <[hidden email]>
>
>
>
>
>
> > Subject: Re: Deconvolution of 3D SIM data
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > When i generate a PSF for deconvolution using suitable beads, then

image=

>
>
>
>
>
> > the same beads again and deconvolve the image, i would expect to get
> real=
>
>
>
>
>
> > ly tiny dots. See e.g.

http://www.svi.nl/BeadsDeconvolutionExample=20

>
>
>
>
>
>
>
>
>
>
>
> real tiny? The images of sub resolution objects dont get very much

smaller

> after
>
>
>
>
>
> deconvolution...
>
>
>
>
>
> what you really get is much higher contrast, so the features look

"sharper"
>
>
>
>
>
> You also get a little bit more resolution.. but thats not really the
main

> point.
>
>
>
>
>
>
>
>
>
>
>
> Its really about contrast.
>
>
>
>
>
>
>
>
>
>
>
> > I would expect something similar for the OMX when the reconstruction

is
> pe=
>
>
>
>
>
> > rfect and includes proper deconvolution. However we get larger
features,=
>
>
>
>
>
> > still within the expected OMX resolution 120 nm in width and 300 nm
in

> z,=
>
>
>
>
>
> > but no dots.
>
>
>
>
>
>
>
>
>
>
>
> No reconstruction or deconvolution will give images that contain

features

> that
>
>
>
>
>
> are smaller
>
>
>
>
>
> than the resolution limit of the system.
>
>
>
>
>
>
>
>
>
>
>
> What do you mean by "dots" ?
>
>
>
>
>
> Single 40 nm pixels?
>
>
>
>
>
>
>
>
>
>
>
> In the OMX it's twice the conventional resolution,
>
>
>
>
>
> so you cant get objects that appear smaller than 120 nm.
>
>
>
>
>
> (actually... API say the OMX system sometimes seems to outperform

theory

>
>
>
>
>
> slightly... but only a bit)
>
>
>
>
>
>
>
>
>
>
>
> You should not an must not get single pixel object with the

reconstructed

> pixel
>
>
>
>
>
> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
> resolution
>
>
>
>
>
> increase from the physical 80 nm pixel spacing)
>
>
>
>
>
>
>
>
>
>
>
> If you get smaller objects, its wrong, and an artifact. No?
>
>
>
>
>
>
>
>
>
>
>
> > I think there is still some improvement possible either in=
>
>
>
>
>
> > the setup of the instrument or the software.
>
>
>
>
>
>
>
>
>
>
>
> Actually, in my hands it seems to do what theory predicts and no more.
>
>
>
>
>
> There is no good reason to expect it to make smaller objects than 120

nm.

>
>
>
>
>
>
>
>
>
>
>
> Certainly the careful alignment of the optics and the highest quality
> objective
>
>
>
>
>
> lens are critical,
>
>
>
>
>
> as is the measurement of suitable SIM PSFs  and careful
> calibration/measurement
>
>
>
>
>
> of the parameters for the reconstruction...
>
>
>
>
>
> the angles and the phases.
>
>
>
>
>
>
>
>
>
>
>
> So, other than making it faster and more tolerant to sub optimal input
> data,
>
>
>
>
>
> I dont think there is any improvement to be made in resolution.. so I

dont

> see
>
>
>
>
>
> how you think there is room for resolution improvement?
>
>
>
>
>
> Maybe you can share your ideas?
>
>
>
>
>
>
>
>
>
>
>
> cheers
>
>
>
>
>
>
>
>
>
>
>
> Dan
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> >
>
>
>
>
>
> > best wishes
>
>
>
>
>
> >
>
>
>
>
>
> > Andreas
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
>
>
>
>
>
>
> 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 )
>
>
>
>
>
>
>
>
>

________________________________

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11

[Guy Cox-2]

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

Abbe considered rays diffracted by two points on the sample.  The points
will be resolved if their diffracted rays can enter the objective.  This
can only apply to  a specimen illuminated from an external source.  In
fluorescence each point emits light and this is totally incoherent with
respect to light from another point.   There is no diffraction at the
sample so Abbe's calculation cannot be applied.  Rayleigh's criterion,
based on how the optics turn the image of a point into a disk (the Airy
disk) does apply.

 

 
Guy

 

Sponsor my next half-marathon on May 15th

There's a special reason - find it out at

http://www.everydayhero.com.au/Guy_Cox_4846
<http://www.everydayhero.com.au/Guy_Cox_4846>

______________________________________________

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 Martin Wessendorf
Sent: Thursday, 14 April 2011 12:10 AM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

 

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

On 4/13/2011 7:33 AM, Guy Cox wrote:

> Abbe's limit,
> on the other hand, is non-negotiable, but it doesn't apply in
> fluorescence.

Guy--can you elaborate a bit on your statement that, "it doesn't apply
in fluorescence"?  Are you talking about STED, or talking in more
general terms?

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]

________________________________

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11

[Guy Cox-2]

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

Yes indeed, so long as your know (eg from TEM) the shape of the
structure you are labelling.  Several studies have done this with great
effect.  Whether this is true super-resolution is something one could
probably debate ad infinitum - let's not go there.  

 

 
Guy

 

Sponsor my next half-marathon on May 15th

There's a special reason - find it out at

http://www.everydayhero.com.au/Guy_Cox_4846
<http://www.everydayhero.com.au/Guy_Cox_4846>

______________________________________________

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 Gustin, Emmanuel [TIBBE]
Sent: Thursday, 14 April 2011 12:19 AM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

 

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

I think that actually, the critical point is that you *know* the beads
(or in the PALM/STORM case, molecules) are sub-resolution in size? The
image reconstruction uses more information than is present in the image,
and this allows you to go below the optical resolution limit. But there
is no way to tell from the image alone what size the beads are.

I guess that theoretically, if you would feed an algorithm with the
knowledge that you have labeled, for example, tubulin and information
about the patterns you expect to see (size of structures, persistence
lengths of bending, branching behavior, ...) you could use it to do
a type of "deconvolution plus" and create a sub-resolution image.

Whether other people would have confidence in the image that has been
constructed in that way, is another matter.

Best Regards,

Emmanuel


--
 Emmanuel Gustin,    Tel. [phone],    e-mail:
[hidden email]  


-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]]
On Behalf Of Guy Cox
Sent: woensdag 13 april 2011 14:34
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data

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

Well, the critical point is the 'sufficiently spaced' bit.  If your
points are well separated you can find their centroids with a precision
way beyond the Rayleigh diffraction limit, which essentially applies to
objects separated by the diffraction limit.  Rayleigh's criterion is
only a rule of thumb - with sufficient signal to noise you can separate
objects closer together than the limit by deconvolution.   Abbe's limit,
on the other hand, is non-negotiable, but it doesn't apply in
fluorescence.  However, when it comes to SIM data, as Nuno has pointed
out, we have already defined the bandwith we are probing, and there is
nothing more to get.




Guy



Sponsor my next half-marathon on May 15th

There's a special reason - find it out at

http://www.everydayhero.com.au/Guy_Cox_4846
<http://www.everydayhero.com.au/Guy_Cox_4846>

______________________________________________

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 Christophe Leterrier
Sent: Wednesday, 13 April 2011 7:31 PM
To: [hidden email]
Subject: Re: Deconvolution of 3D SIM data



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

To elaborate a bit on that, why isn't it possible to go beyond the
diffraction limit with deconvolution ? I don't see the theoretical
reason,
in a ideal case (subresolutive beads sufficiently spaced). Of course in
a
real sample, S/N issues and fluorophore density impose a limit on the
precision, but why would this limit precisely be at the diffraction
limit ?

Christophe

On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]>
wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
>
> Hi Dan, Brian
>
> thanks for the comments, i know that for real live samples everything
what
> exceeds the resolution limit of the system is an artefact, however for
> sparsely separated beads i would expect that i can determine their
position
> with high accuracy only limited by the signal to noise ratio. This is
> exactly what is done in localisation microscopy (PALM, STORM...). One
> approach is to calculate the cross correlation between the measured 3D
PSF
> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011,
327).
> While this is not exactly what deconvolution does, it is at least
related to
> deconvolution.
>
> So i was wondering if this special case, a limited number of well
resolved
> beads which are small enough to qualify as point objects, could tell
us how
> good the deconvolution method actually works, or in the case of SIM,
what
> the image restauration actually does. At least theoretically, the size
of
> the beads (110 nm in my case) should not be a problem as by using the
same
> beads to measure the PSF we trick the deconvolution into thinking that
they
> are point objects. This is similar to the obove mentioned localisation
> microscopy which does not depend much on optical resolution or pixel
size.

>
> best wishes
>
> Andreas
>
>
>
>
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: Daniel James White <[hidden email]>
> To: [hidden email]
> Sent: Mon, 11 Apr 2011 10:53
> Subject: Re: Deconvolution of 3D SIM data
>
>
> *****
>
>
>
>
>
> To join, leave or search the confocal microscopy listserv, go to:
>
>
>
>
>
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>
>
>
>
>
> *****
>
>
>
>
>
>
>
>
>
>
>
> Hi Andreas,
>
>
>
>
>
>
>
>
>
>
>
> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
> wrote:
>
>
>
>
>
>
>
>
>
>
>
> > Date:    Fri, 8 Apr 2011 11:48:46 -0400
>
>
>
>
>
> > From:    Andreas Bruckbauer <[hidden email]>
>
>
>
>
>
> > Subject: Re: Deconvolution of 3D SIM data
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > When i generate a PSF for deconvolution using suitable beads, then

image=

>
>
>
>
>
> > the same beads again and deconvolve the image, i would expect to get
> real=
>
>
>
>
>
> > ly tiny dots. See e.g.

http://www.svi.nl/BeadsDeconvolutionExample=20

>
>
>
>
>
>
>
>
>
>
>
> real tiny? The images of sub resolution objects dont get very much

smaller

> after
>
>
>
>
>
> deconvolution...
>
>
>
>
>
> what you really get is much higher contrast, so the features look

"sharper"
>
>
>
>
>
> You also get a little bit more resolution.. but thats not really the
main

> point.
>
>
>
>
>
>
>
>
>
>
>
> Its really about contrast.
>
>
>
>
>
>
>
>
>
>
>
> > I would expect something similar for the OMX when the reconstruction

is
> pe=
>
>
>
>
>
> > rfect and includes proper deconvolution. However we get larger
features,=
>
>
>
>
>
> > still within the expected OMX resolution 120 nm in width and 300 nm
in

> z,=
>
>
>
>
>
> > but no dots.
>
>
>
>
>
>
>
>
>
>
>
> No reconstruction or deconvolution will give images that contain

features

> that
>
>
>
>
>
> are smaller
>
>
>
>
>
> than the resolution limit of the system.
>
>
>
>
>
>
>
>
>
>
>
> What do you mean by "dots" ?
>
>
>
>
>
> Single 40 nm pixels?
>
>
>
>
>
>
>
>
>
>
>
> In the OMX it's twice the conventional resolution,
>
>
>
>
>
> so you cant get objects that appear smaller than 120 nm.
>
>
>
>
>
> (actually... API say the OMX system sometimes seems to outperform

theory

>
>
>
>
>
> slightly... but only a bit)
>
>
>
>
>
>
>
>
>
>
>
> You should not an must not get single pixel object with the

reconstructed

> pixel
>
>
>
>
>
> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
> resolution
>
>
>
>
>
> increase from the physical 80 nm pixel spacing)
>
>
>
>
>
>
>
>
>
>
>
> If you get smaller objects, its wrong, and an artifact. No?
>
>
>
>
>
>
>
>
>
>
>
> > I think there is still some improvement possible either in=
>
>
>
>
>
> > the setup of the instrument or the software.
>
>
>
>
>
>
>
>
>
>
>
> Actually, in my hands it seems to do what theory predicts and no more.
>
>
>
>
>
> There is no good reason to expect it to make smaller objects than 120

nm.

>
>
>
>
>
>
>
>
>
>
>
> Certainly the careful alignment of the optics and the highest quality
> objective
>
>
>
>
>
> lens are critical,
>
>
>
>
>
> as is the measurement of suitable SIM PSFs  and careful
> calibration/measurement
>
>
>
>
>
> of the parameters for the reconstruction...
>
>
>
>
>
> the angles and the phases.
>
>
>
>
>
>
>
>
>
>
>
> So, other than making it faster and more tolerant to sub optimal input
> data,
>
>
>
>
>
> I dont think there is any improvement to be made in resolution.. so I

dont

> see
>
>
>
>
>
> how you think there is room for resolution improvement?
>
>
>
>
>
> Maybe you can share your ideas?
>
>
>
>
>
>
>
>
>
>
>
> cheers
>
>
>
>
>
>
>
>
>
>
>
> Dan
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> >
>
>
>
>
>
> > best wishes
>
>
>
>
>
> >
>
>
>
>
>
> > Andreas
>
>
>
>
>
> >
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
> >
>
>
>
>
>
> > =20
>
>
>
>
>
>
>
>
>
>
>
> 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 )
>
>
>
>
>
>
>
>
>

________________________________

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11

________________________________

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11

[Cory Quammen]

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

> Yes indeed, so long as your know (eg from TEM) the shape of the
> structure you are labelling.  Several studies have done this with great
> effect.  Whether this is true super-resolution is something one could
> probably debate ad infinitum - let's not go there.
>

Guy,

Could you give some pointers to these studies? I'm quite interested in
the approach.

Thanks,
Cory

>
> Sponsor my next half-marathon on May 15th
>
> There's a special reason - find it out at
>
> http://www.everydayhero.com.au/Guy_Cox_4846
> <http://www.everydayhero.com.au/Guy_Cox_4846>
>
> ______________________________________________
>
> 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 Gustin, Emmanuel [TIBBE]
> Sent: Thursday, 14 April 2011 12:19 AM
> To: [hidden email]
> Subject: Re: Deconvolution of 3D SIM data
>
>
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> I think that actually, the critical point is that you *know* the beads
> (or in the PALM/STORM case, molecules) are sub-resolution in size? The
> image reconstruction uses more information than is present in the image,
> and this allows you to go below the optical resolution limit. But there
> is no way to tell from the image alone what size the beads are.
>
> I guess that theoretically, if you would feed an algorithm with the
> knowledge that you have labeled, for example, tubulin and information
> about the patterns you expect to see (size of structures, persistence
> lengths of bending, branching behavior, ...) you could use it to do
> a type of "deconvolution plus" and create a sub-resolution image.
>
> Whether other people would have confidence in the image that has been
> constructed in that way, is another matter.
>
> Best Regards,
>
> Emmanuel
>
>
> --
>  Emmanuel Gustin,    Tel. [phone],    e-mail:
> [hidden email]
>
>
> -----Original Message-----
> From: Confocal Microscopy List [mailto:[hidden email]]
> On Behalf Of Guy Cox
> Sent: woensdag 13 april 2011 14:34
> To: [hidden email]
> Subject: Re: Deconvolution of 3D SIM data
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Well, the critical point is the 'sufficiently spaced' bit.  If your
> points are well separated you can find their centroids with a precision
> way beyond the Rayleigh diffraction limit, which essentially applies to
> objects separated by the diffraction limit.  Rayleigh's criterion is
> only a rule of thumb - with sufficient signal to noise you can separate
> objects closer together than the limit by deconvolution.   Abbe's limit,
> on the other hand, is non-negotiable, but it doesn't apply in
> fluorescence.  However, when it comes to SIM data, as Nuno has pointed
> out, we have already defined the bandwith we are probing, and there is
> nothing more to get.
>
>
>
>
> Guy
>
>
>
> Sponsor my next half-marathon on May 15th
>
> There's a special reason - find it out at
>
> http://www.everydayhero.com.au/Guy_Cox_4846
> <http://www.everydayhero.com.au/Guy_Cox_4846>
>
> ______________________________________________
>
> 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 Christophe Leterrier
> Sent: Wednesday, 13 April 2011 7:31 PM
> To: [hidden email]
> Subject: Re: Deconvolution of 3D SIM data
>
>
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> To elaborate a bit on that, why isn't it possible to go beyond the
> diffraction limit with deconvolution ? I don't see the theoretical
> reason,
> in a ideal case (subresolutive beads sufficiently spaced). Of course in
> a
> real sample, S/N issues and fluorophore density impose a limit on the
> precision, but why would this limit precisely be at the diffraction
> limit ?
>
> Christophe
>
> On Wed, Apr 13, 2011 at 09:44, Andreas Bruckbauer <[hidden email]>
> wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>>
>>
>> Hi Dan, Brian
>>
>> thanks for the comments, i know that for real live samples everything
> what
>> exceeds the resolution limit of the system is an artefact, however for
>> sparsely separated beads i would expect that i can determine their
> position
>> with high accuracy only limited by the signal to noise ratio. This is
>> exactly what is done in localisation microscopy (PALM, STORM...). One
>> approach is to calculate the cross correlation between the measured 3D
> PSF
>> and the 3D image (see e.g. A. G. York et al. Nature Methods 8, 2011,
> 327).
>> While this is not exactly what deconvolution does, it is at least
> related to
>> deconvolution.
>>
>> So i was wondering if this special case, a limited number of well
> resolved
>> beads which are small enough to qualify as point objects, could tell
> us how
>> good the deconvolution method actually works, or in the case of SIM,
> what
>> the image restauration actually does. At least theoretically, the size
> of
>> the beads (110 nm in my case) should not be a problem as by using the
> same
>> beads to measure the PSF we trick the deconvolution into thinking that
> they
>> are point objects. This is similar to the obove mentioned localisation
>> microscopy which does not depend much on optical resolution or pixel
> size.
>>
>> best wishes
>>
>> Andreas
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> -----Original Message-----
>> From: Daniel James White <[hidden email]>
>> To: [hidden email]
>> Sent: Mon, 11 Apr 2011 10:53
>> Subject: Re: Deconvolution of 3D SIM data
>>
>>
>> *****
>>
>>
>>
>>
>>
>> To join, leave or search the confocal microscopy listserv, go to:
>>
>>
>>
>>
>>
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>
>>
>>
>>
>>
>> *****
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Hi Andreas,
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> On Apr 9, 2011, at 7:02 AM, CONFOCALMICROSCOPY automatic digest system
>> wrote:
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> > Date:    Fri, 8 Apr 2011 11:48:46 -0400
>>
>>
>>
>>
>>
>> > From:    Andreas Bruckbauer <[hidden email]>
>>
>>
>>
>>
>>
>> > Subject: Re: Deconvolution of 3D SIM data
>>
>>
>>
>>
>>
>> >
>>
>>
>>
>>
>>
>> >
>>
>>
>>
>>
>>
>> > When i generate a PSF for deconvolution using suitable beads, then
> image=
>>
>>
>>
>>
>>
>> > the same beads again and deconvolve the image, i would expect to get
>> real=
>>
>>
>>
>>
>>
>> > ly tiny dots. See e.g.
> http://www.svi.nl/BeadsDeconvolutionExample=20
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> real tiny? The images of sub resolution objects dont get very much
> smaller
>> after
>>
>>
>>
>>
>>
>> deconvolution...
>>
>>
>>
>>
>>
>> what you really get is much higher contrast, so the features look
> "sharper"
>>
>>
>>
>>
>>
>> You also get a little bit more resolution.. but thats not really the
> main
>> point.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Its really about contrast.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> > I would expect something similar for the OMX when the reconstruction
> is
>> pe=
>>
>>
>>
>>
>>
>> > rfect and includes proper deconvolution. However we get larger
> features,=
>>
>>
>>
>>
>>
>> > still within the expected OMX resolution 120 nm in width and 300 nm
> in
>> z,=
>>
>>
>>
>>
>>
>> > but no dots.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> No reconstruction or deconvolution will give images that contain
> features
>> that
>>
>>
>>
>>
>>
>> are smaller
>>
>>
>>
>>
>>
>> than the resolution limit of the system.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> What do you mean by "dots" ?
>>
>>
>>
>>
>>
>> Single 40 nm pixels?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> In the OMX it's twice the conventional resolution,
>>
>>
>>
>>
>>
>> so you cant get objects that appear smaller than 120 nm.
>>
>>
>>
>>
>>
>> (actually... API say the OMX system sometimes seems to outperform
> theory
>>
>>
>>
>>
>>
>> slightly... but only a bit)
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> You should not an must not get single pixel object with the
> reconstructed
>> pixel
>>
>>
>>
>>
>>
>> spacing of 40 nm (thats the spacing we have with the EM CCDs after 2x
>> resolution
>>
>>
>>
>>
>>
>> increase from the physical 80 nm pixel spacing)
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> If you get smaller objects, its wrong, and an artifact. No?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> > I think there is still some improvement possible either in=
>>
>>
>>
>>
>>
>> > the setup of the instrument or the software.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Actually, in my hands it seems to do what theory predicts and no more.
>>
>>
>>
>>
>>
>> There is no good reason to expect it to make smaller objects than 120
> nm.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Certainly the careful alignment of the optics and the highest quality
>> objective
>>
>>
>>
>>
>>
>> lens are critical,
>>
>>
>>
>>
>>
>> as is the measurement of suitable SIM PSFs  and careful
>> calibration/measurement
>>
>>
>>
>>
>>
>> of the parameters for the reconstruction...
>>
>>
>>
>>
>>
>> the angles and the phases.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> So, other than making it faster and more tolerant to sub optimal input
>> data,
>>
>>
>>
>>
>>
>> I dont think there is any improvement to be made in resolution.. so I
> dont
>> see
>>
>>
>>
>>
>>
>> how you think there is room for resolution improvement?
>>
>>
>>
>>
>>
>> Maybe you can share your ideas?
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> cheers
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Dan
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> >
>>
>>
>>
>>
>>
>> > best wishes
>>
>>
>>
>>
>>
>> >
>>
>>
>>
>>
>>
>> > Andreas
>>
>>
>>
>>
>>
>> >
>>
>>
>>
>>
>>
>> >
>>
>>
>>
>>
>>
>> > =20
>>
>>
>>
>>
>>
>> >
>>
>>
>>
>>
>>
>> > =20
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> 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 )
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
> ________________________________
>
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11
>
> ________________________________
>
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 10.0.1209 / Virus Database: 1500/3569 - Release Date: 04/12/11
>