signal to background ratios in multiphoton microscopy

[Michael Giacomelli]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

I have been reading up on signal to background ratios in point
scanning confocal and spinning disk confocal.   There are a number of
very detailed papers that characterize the out of focus rejection in
terms of the edge response function for a thick fluorescent sample in
the one photon case (figure 14, [1] for instance) .

What i have not been able to locate is a reasonably thorough treatment
of the subject for multiphoton microscopy with non-descanned
detection.  Instead most of the work I can find only looks at the
signal to background ratios with depth into a tissue specimen.  This
is helpful, but difficult to compare since you can't image deeply with
confocal.

Does anyone have a good reference they could suggest?


[1]  Zimmermann, B., Kempe, M., & Wolleschensky, R. (2006). High-speed
confocal fluorescence imaging with a novel line scanning microscope.
Journal of Biomedical Optics, 11(6), 064011.
http://doi.org/10.1117/1.2402110

[Craig Brideau]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Because the signal is only generated at the focal plane, multiphoton
doesn't have the same kind of background issues as confocal. This means
there is no out-of-focus signal to reject. The spatial restriction means
that your laser's point spread function (in XYZ) is the limit of your
detection and your signal generation volume. For instance in the edge
response to a thick fluorescent sample, you won't get any signal generated
until the PSF of your laser actually starts overlapping with the edge of
the slab. Conversely in confocal, you will always have signal being
generated and are relying on the pinhole to reject the out-of-plane light.
As you take a volume stack into a slab In 2P you will get some entertaining
edge effects due to spherical aberration if the index of your fluorescent
slab isn't matched to your immersion media, but otherwise 2P just doesn't
generate signal unless the sample is within the focal spot. Please keep in
mind I am thinking about the Z direction in this, not XY.

I hope this helps; I'm really tired and may not be as coherent as usual. :)
Craig

On Wed, May 11, 2016 at 3:01 PM, Michael Giacomelli <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> I have been reading up on signal to background ratios in point
> scanning confocal and spinning disk confocal.   There are a number of
> very detailed papers that characterize the out of focus rejection in
> terms of the edge response function for a thick fluorescent sample in
> the one photon case (figure 14, [1] for instance) .
>
> What i have not been able to locate is a reasonably thorough treatment
> of the subject for multiphoton microscopy with non-descanned
> detection.  Instead most of the work I can find only looks at the
> signal to background ratios with depth into a tissue specimen.  This
> is helpful, but difficult to compare since you can't image deeply with
> confocal.
>
> Does anyone have a good reference they could suggest?
>
>
> [1]  Zimmermann, B., Kempe, M., & Wolleschensky, R. (2006). High-speed
> confocal fluorescence imaging with a novel line scanning microscope.
> Journal of Biomedical Optics, 11(6), 064011.
> http://doi.org/10.1117/1.2402110
>

[Michael Giacomelli]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Craig,

I realize that 2 photon excitation is most likely in the focal plane,
but the probability of it happening is nonzero everywhere the beam
reaches, so you do get some signal from out of focus light as in the
confocal case.  This is an issue in deep brain imaging for instance.
I spent a little while searching on Google scholar but didn't see very
much on signal to background ratios derived or measured for the step
test though, just deep tissue imaging.

For what it's worth, I'm curious because I do see some depth dependent
background when Z-stacking the air above a fluorescent slide.  It's
weak, but I'm curious if what I measure is reasonable, or if I should
be double checking my PSF and filter rejection.

Mike

On Wed, May 11, 2016 at 9:01 PM, Craig Brideau <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Because the signal is only generated at the focal plane, multiphoton
> doesn't have the same kind of background issues as confocal. This means
> there is no out-of-focus signal to reject. The spatial restriction means
> that your laser's point spread function (in XYZ) is the limit of your
> detection and your signal generation volume. For instance in the edge
> response to a thick fluorescent sample, you won't get any signal generated
> until the PSF of your laser actually starts overlapping with the edge of
> the slab. Conversely in confocal, you will always have signal being
> generated and are relying on the pinhole to reject the out-of-plane light.
> As you take a volume stack into a slab In 2P you will get some entertaining
> edge effects due to spherical aberration if the index of your fluorescent
> slab isn't matched to your immersion media, but otherwise 2P just doesn't
> generate signal unless the sample is within the focal spot. Please keep in
> mind I am thinking about the Z direction in this, not XY.
>
> I hope this helps; I'm really tired and may not be as coherent as usual. :)
> Craig
>
> On Wed, May 11, 2016 at 3:01 PM, Michael Giacomelli <[hidden email]> wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> Post images on http://www.imgur.com and include the link in your posting.
>> *****
>>
>> I have been reading up on signal to background ratios in point
>> scanning confocal and spinning disk confocal.   There are a number of
>> very detailed papers that characterize the out of focus rejection in
>> terms of the edge response function for a thick fluorescent sample in
>> the one photon case (figure 14, [1] for instance) .
>>
>> What i have not been able to locate is a reasonably thorough treatment
>> of the subject for multiphoton microscopy with non-descanned
>> detection.  Instead most of the work I can find only looks at the
>> signal to background ratios with depth into a tissue specimen.  This
>> is helpful, but difficult to compare since you can't image deeply with
>> confocal.
>>
>> Does anyone have a good reference they could suggest?
>>
>>
>> [1]  Zimmermann, B., Kempe, M., & Wolleschensky, R. (2006). High-speed
>> confocal fluorescence imaging with a novel line scanning microscope.
>> Journal of Biomedical Optics, 11(6), 064011.
>> http://doi.org/10.1117/1.2402110
>>

[Michael Giacomelli]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

After some more reading, I found a reference:

http://onlinelibrary.wiley.com/doi/10.1046/j.1365-2818.1998.00431.x/abstract

My edge response looks reasonable, but the long tail I get is a little
too high, so something isn't quite right.

Mike

On Wed, May 11, 2016 at 9:31 PM, Michael Giacomelli <[hidden email]> wrote:

> Hi Craig,
>
> I realize that 2 photon excitation is most likely in the focal plane,
> but the probability of it happening is nonzero everywhere the beam
> reaches, so you do get some signal from out of focus light as in the
> confocal case.  This is an issue in deep brain imaging for instance.
> I spent a little while searching on Google scholar but didn't see very
> much on signal to background ratios derived or measured for the step
> test though, just deep tissue imaging.
>
> For what it's worth, I'm curious because I do see some depth dependent
> background when Z-stacking the air above a fluorescent slide.  It's
> weak, but I'm curious if what I measure is reasonable, or if I should
> be double checking my PSF and filter rejection.
>
> Mike
>
> On Wed, May 11, 2016 at 9:01 PM, Craig Brideau <[hidden email]> wrote:
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> Post images on http://www.imgur.com and include the link in your posting.
>> *****
>>
>> Because the signal is only generated at the focal plane, multiphoton
>> doesn't have the same kind of background issues as confocal. This means
>> there is no out-of-focus signal to reject. The spatial restriction means
>> that your laser's point spread function (in XYZ) is the limit of your
>> detection and your signal generation volume. For instance in the edge
>> response to a thick fluorescent sample, you won't get any signal generated
>> until the PSF of your laser actually starts overlapping with the edge of
>> the slab. Conversely in confocal, you will always have signal being
>> generated and are relying on the pinhole to reject the out-of-plane light.
>> As you take a volume stack into a slab In 2P you will get some entertaining
>> edge effects due to spherical aberration if the index of your fluorescent
>> slab isn't matched to your immersion media, but otherwise 2P just doesn't
>> generate signal unless the sample is within the focal spot. Please keep in
>> mind I am thinking about the Z direction in this, not XY.
>>
>> I hope this helps; I'm really tired and may not be as coherent as usual. :)
>> Craig
>>
>> On Wed, May 11, 2016 at 3:01 PM, Michael Giacomelli <[hidden email]> wrote:
>>
>>> *****
>>> To join, leave or search the confocal microscopy listserv, go to:
>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>> Post images on http://www.imgur.com and include the link in your posting.
>>> *****
>>>
>>> I have been reading up on signal to background ratios in point
>>> scanning confocal and spinning disk confocal.   There are a number of
>>> very detailed papers that characterize the out of focus rejection in
>>> terms of the edge response function for a thick fluorescent sample in
>>> the one photon case (figure 14, [1] for instance) .
>>>
>>> What i have not been able to locate is a reasonably thorough treatment
>>> of the subject for multiphoton microscopy with non-descanned
>>> detection.  Instead most of the work I can find only looks at the
>>> signal to background ratios with depth into a tissue specimen.  This
>>> is helpful, but difficult to compare since you can't image deeply with
>>> confocal.
>>>
>>> Does anyone have a good reference they could suggest?
>>>
>>>
>>> [1]  Zimmermann, B., Kempe, M., & Wolleschensky, R. (2006). High-speed
>>> confocal fluorescence imaging with a novel line scanning microscope.
>>> Journal of Biomedical Optics, 11(6), 064011.
>>> http://doi.org/10.1117/1.2402110
>>>

[Craig Brideau]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

If you are referring to a plot of Z depth to signal intensity, it will be
very asymmetric due to spherical aberration caused as you go into the
interface. I have also achieved strange results when doing this myself by
saturating all the fluorophores within the PSF which will give you very
weird Z vs intensity plots. Try keeping your excitation power low to avoid
this.

Craig
On May 11, 2016 8:30 PM, "Michael Giacomelli" <[hidden email]> wrote:

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

After some more reading, I found a reference:

http://onlinelibrary.wiley.com/doi/10.1046/j.1365-2818.1998.00431.x/abstract

My edge response looks reasonable, but the long tail I get is a little
too high, so something isn't quite right.

Mike

On Wed, May 11, 2016 at 9:31 PM, Michael Giacomelli <[hidden email]> wrote:

> Hi Craig,
>
> I realize that 2 photon excitation is most likely in the focal plane,
> but the probability of it happening is nonzero everywhere the beam
> reaches, so you do get some signal from out of focus light as in the
> confocal case.  This is an issue in deep brain imaging for instance.
> I spent a little while searching on Google scholar but didn't see very
> much on signal to background ratios derived or measured for the step
> test though, just deep tissue imaging.
>
> For what it's worth, I'm curious because I do see some depth dependent
> background when Z-stacking the air above a fluorescent slide.  It's
> weak, but I'm curious if what I measure is reasonable, or if I should
> be double checking my PSF and filter rejection.
>
> Mike
>
> On Wed, May 11, 2016 at 9:01 PM, Craig Brideau <[hidden email]>

wrote:

>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> Post images on http://www.imgur.com and include the link in your posting.
>> *****
>>
>> Because the signal is only generated at the focal plane, multiphoton
>> doesn't have the same kind of background issues as confocal. This means
>> there is no out-of-focus signal to reject. The spatial restriction means
>> that your laser's point spread function (in XYZ) is the limit of your
>> detection and your signal generation volume. For instance in the edge
>> response to a thick fluorescent sample, you won't get any signal

generated
>> until the PSF of your laser actually starts overlapping with the edge of
>> the slab. Conversely in confocal, you will always have signal being
>> generated and are relying on the pinhole to reject the out-of-plane
light.
>> As you take a volume stack into a slab In 2P you will get some
entertaining
>> edge effects due to spherical aberration if the index of your fluorescent
>> slab isn't matched to your immersion media, but otherwise 2P just doesn't
>> generate signal unless the sample is within the focal spot. Please keep
in
>> mind I am thinking about the Z direction in this, not XY.
>>
>> I hope this helps; I'm really tired and may not be as coherent as usual.
:)
>> Craig
>>
>> On Wed, May 11, 2016 at 3:01 PM, Michael Giacomelli <[hidden email]> wrote:
>>
>>> *****
>>> To join, leave or search the confocal microscopy listserv, go to:
>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>> Post images on http://www.imgur.com and include the link in your
posting.

>>> *****
>>>
>>> I have been reading up on signal to background ratios in point
>>> scanning confocal and spinning disk confocal.   There are a number of
>>> very detailed papers that characterize the out of focus rejection in
>>> terms of the edge response function for a thick fluorescent sample in
>>> the one photon case (figure 14, [1] for instance) .
>>>
>>> What i have not been able to locate is a reasonably thorough treatment
>>> of the subject for multiphoton microscopy with non-descanned
>>> detection.  Instead most of the work I can find only looks at the
>>> signal to background ratios with depth into a tissue specimen.  This
>>> is helpful, but difficult to compare since you can't image deeply with
>>> confocal.
>>>
>>> Does anyone have a good reference they could suggest?
>>>
>>>
>>> [1]  Zimmermann, B., Kempe, M., & Wolleschensky, R. (2006). High-speed
>>> confocal fluorescence imaging with a novel line scanning microscope.
>>> Journal of Biomedical Optics, 11(6), 064011.
>>> http://doi.org/10.1117/1.2402110
>>>