Two Photon Microscope Low Signal Troubleshooting
Locked
 
|
Heping Yuan |
Two Photon Microscope Low Signal Troubleshooting
|
*****
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, I hope the mailing list doesn't mind being bothered with another question. I am currently setting up a new custom resonant scanning two photon system (old Spectra-Physics Mai Tai, 8kHz resonant scanners from CamTech and new H10770B-40 PMT's from Hamamatsu). The problem I'm having is that when I attempt to do in vivo calcium imaging (GCAMP6s), it seems that my signal is too weak / SNR is too low when I try to keep the laser power after the objective below what I understand to be safe levels (< 50 mW). I have checked the alignment of the system multiple times including: 1) Beam is collimated and centered prior to Galvo's 1) Beam is centered and collimated after the beam expander following the Galvo's 2) Beam is centered at the back aperture of the objective 3) PMT collection lens position is adjusted for maximum PMT signal The entire system is inside a cage system so alignment should be very close to begin with. Most parameters such as diameter of beam at back aperture of objective I have taken from what others have used in literature (18.5 mm for 20x Olympus objective) I understand there could be many factors that contribute to this problem (such as GCAMP expression, cranial window quality on the sample side) although I have tested on multiple mice. I guess I'm wondering if there are tests people use for troubleshooting a two photon microscope with lower than expected signal that are beyond the kind of alignment tests I mentioned above. Are there performance standards I can use (ex: for a given laser power after the objective and a given sample, one should be able to obtain a certain SNR) so that I have some idea how optimized my system currently is? One thing we don't have in our setup is any kind of dispersion compensation. Could this alone be enough to allow us to image with sufficient SNR at lower powers? Is it possible that the beam quality of the laser could result in low image SNR (aside from dispersion effects) and would it be something we could check? Any help would be greatly appreciated, Tim |
|
|
Michael Giacomelli |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
*****
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 Tim, Which Mai Tai are you using and what is the center wavelength you've tuned it to? I think there are some fairly short pulse length variants of that laser which would benefit from pulse compression, but the normal ~100 fs variant is probably ok with it if the optics in your system are dispersion optimized (e.g. not too many achromats, etc). Do you have a sense if the optical design is reasonably good? Its usually pretty easy to spot problems like astigmatic scanning just by scanning a uniform fluorescent solution, but things like uncorrected spherical or chromatic aberration can be maddeningly difficult to diagnose without careful PSF measurements. Mike On Wed, Sep 28, 2016 at 11:28 PM, Tim <[hidden email]> wrote: ***** |
|
|
Craig Brideau |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
*****
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.
*****
An easy check is to measure the power at the sample plane. You might have unexpected losses in the system that are reducing power at the sample. Craig On Sep 28, 2016 9:55 PM, "Michael Giacomelli" <[hidden email]> wrote: ***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/ |
|
|
0000001ed7f52e4a-dmarc-request |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
In reply to this post by Heping Yuan
*****
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 Tim, Just put some green PS Speck beads on a slide and image at 800 nm, when the image is not too far off the expected values, your system should be fine. If you have a convallaria test sample, this should give a bright signal at 800 nm with just a few mW and low detector gain. How deep into the tissue are you trying to image? It might be good to underfill the back aperture to gain signal at the expense of z-resolution. Best wishes Andreas From: [hidden email] Sent: 29/09/2016 04:28 To: [hidden email] Subject: Two Photon Microscope Low Signal Troubleshooting 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, I hope the mailing list doesn't mind being bothered with another question. I am currently setting up a new custom resonant scanning two photon system (old Spectra-Physics Mai Tai, 8kHz resonant scanners from CamTech and new H10770B-40 PMT's from Hamamatsu). The problem I'm having is that when I attempt to do in vivo calcium imaging (GCAMP6s), it seems that my signal is too weak / SNR is too low when I try to keep the laser power after the objective below what I understand to be safe levels (< 50 mW). I have checked the alignment of the system multiple times including: 1) Beam is collimated and centered prior to Galvo's 1) Beam is centered and collimated after the beam expander following the Galvo's 2) Beam is centered at the back aperture of the objective 3) PMT collection lens position is adjusted for maximum PMT signal The entire system is inside a cage system so alignment should be very close to begin with. Most parameters such as diameter of beam at back aperture of objective I have taken from what others have used in literature (18.5 mm for 20x Olympus objective) I understand there could be many factors that contribute to this problem (such as GCAMP expression, cranial window quality on the sample side) although I have tested on multiple mice. I guess I'm wondering if there are tests people use for troubleshooting a two photon microscope with lower than expected signal that are beyond the kind of alignment tests I mentioned above. Are there performance standards I can use (ex: for a given laser power after the objective and a given sample, one should be able to obtain a certain SNR) so that I have some idea how optimized my system currently is? One thing we don't have in our setup is any kind of dispersion compensation. Could this alone be enough to allow us to image with sufficient SNR at lower powers? Is it possible that the beam quality of the laser could result in low image SNR (aside from dispersion effects) and would it be something we could check? Any help would be greatly appreciated, Tim |
|
|
Heping Yuan |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
In reply to this post by Heping Yuan
*****
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. ***** Thank you all for your responses. @ Michael: Our Ti-Sapphire has 100 fs pulses and is tuned to 900 nm. We use only protected silver mirrors and the following lenses from Thorlabs: Initial Beam expansion with: https://www.thorlabs.com/thorproduct.cfm?partnumber=AC254-030-B https://www.thorlabs.com/thorproduct.cfm?partnumber=AC254-060-B Scan lens: https://www.thorlabs.com/thorproduct.cfm?partnumber=LSM05-BB Tube lens: https://www.thorlabs.com/thorproduct.cfm?partnumber=AC508-400-B Do you think this would this constitute too many achromats? @ Craig: I am measuring power after objective using a power meter from Thorlabs (https://www.thorlabs.com/thorproduct.cfm?partnumber=S121C) held up against output end of the objective. @ Andreas: For the PS Speck beads on a slide, when you say that the image is not too far off the expected values do you mean the expected spatial resolution values of the system? |
|
|
0000001ed7f52e4a-dmarc-request |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
*****
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.
*****
@ Andreas: For the PS Speck beads on a slide, when you say that the image is not too far off the expected values do you mean the expected spatial resolution values of the system? Yes, I get about 410 nm fwhm in xy and about 1.2 um in z with a 25x 1.05 NA objective when the back aperture is nicely filled. If the z resolution is bad, you might not see small objects but will do just fine imaging larger ones. Also check if you are bleaching the sample before you actually see it, but in this case I would expect a short spike in fluorescence. Best wishes Andreas From: [hidden email] Sent: 29/09/2016 07:25 To: [hidden email] Subject: Re: Two Photon Microscope Low Signal Troubleshooting 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. ***** Thank you all for your responses. @ Michael: Our Ti-Sapphire has 100 fs pulses and is tuned to 900 nm. We use only protected silver mirrors and the following lenses from Thorlabs: Initial Beam expansion with: https://www.thorlabs.com/thorproduct.cfm?partnumber=AC254-030-B https://www.thorlabs.com/thorproduct.cfm?partnumber=AC254-060-B Scan lens: https://www.thorlabs.com/thorproduct.cfm?partnumber=LSM05-BB Tube lens: https://www.thorlabs.com/thorproduct.cfm?partnumber=AC508-400-B Do you think this would this constitute too many achromats? @ Craig: I am measuring power after objective using a power meter from Thorlabs (https://www.thorlabs.com/thorproduct.cfm?partnumber=S121C) held up against output end of the objective. @ Andreas: For the PS Speck beads on a slide, when you say that the image is not too far off the expected values do you mean the expected spatial resolution values of the system? |
|
|
Zdenek Svindrych-2 |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
In reply to this post by Heping Yuan
*****
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 Tim,
nice scan lens! I was always wondering whether that's necessary, it seems like an overkill for me. I though an achromatic doublet would do a good job, especially in intravital imaging, where the flatness of the field is not so critical (the mouse brain is not flat, after all). Some people use beam cleanup before the scanner to make sure they're using just the fundamental mode for excitation. But I can't comment on the benefits in terms of reducing the incident laser power while maintaining the fluorescence intensity. In theory, if the laser is good (low M^2) and all mirrors flat, it shouldn't be necessary. Another place where you can loose a lot of light is the detectors. Are you sure you're collecting all the light the objective can gather? The PMT's you're using have much smaller effective area then the old multialkali tubes (5 mm diameter vs maybe 20 mm), so you want to make sure your optical design is perfect! I've spent a number of sleepless nights trying to image the BFP aperture of my objective (just over 10 mm dia) onto my detectors (R10467U inside HPM-100 module, just 3 mm diameter sensitive area)... Good luck! zdenek -- Zdenek Svindrych, Ph.D. W.M. Keck Center for Cellular Imaging (PLSB 003) University of Virginia, Charlottesville, VA http://www.kcci.virginia.edu/ tel: 434-982-4869 Annual FRET Workshop: http://kcci.virginia.edu/workshop-2017 ---------- Původní zpráva ---------- ***** |
|
|
Craig Brideau |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
*****
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 always put a strong aspheric lens directly in front of my PMT to increase its effective aperture. You can get away with an aspheric condenser rather than an achromat at that point, and aspheres can have extremely high NA. Craig On Sep 29, 2016 8:33 AM, <[hidden email]> wrote: ***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/ |
|
|
Michael Giacomelli |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
In reply to this post by Heping Yuan
*****
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 Tim, Those achromats are extremely dispersive. That 30 mm is contributing 1100 fs^2 at 900 alone. Are you really sure you need all of them? Designing a low dispersion beam expander without using achromats is simple using only Thorlabs stock singles for example. Usually dispersion is not a huge factor at 100 fs, but you still probably want to be very careful about using achromats except where absolutely necessary. They're typically about 5-10x more dispersive then equivalent singlets, and its not too hard to find single parts in the Thor catalog that have more dispersion than typical objectives. By the way, if you haven't checked it out, you can look up the dispersion of stock optical glasses here: Might be worth calculating how much pulse broadening you are getting from your optics. Mike On Thu, Sep 29, 2016 at 2:25 AM, Tim <[hidden email]> wrote: ***** |
|
|
Needham, Maurice |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
In reply to this post by 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.
*****
Hi Tim, as you mentioned, there may not be a problem with your set-up at all. We do imaging through cranial windows using resonant 2-photon with a coherent chameleon II laser. We have found much variety in the strain of GCaMP6 mouse used where some signals are stronger than others. Of course this requires keeping the GCaMP6 variant consistent (slow, medium, fast). Our transgenic samples are dimmer than vector injected samples and we have tried multiple GCaMP6 transgenics with variation in their signals also. We currently use around 80mW at the objective so consider increasing the laser power and if you are causing no damage then you are fine.
Maurice From: Confocal Microscopy List <[hidden email]> on behalf of Craig Brideau <[hidden email]>
Sent: September 29, 2016 9:38:24 AM To: [hidden email] Subject: Re: Two Photon Microscope Low Signal Troubleshooting ***** 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 always put a strong aspheric lens directly in front of my PMT to increase its effective aperture. You can get away with an aspheric condenser rather than an achromat at that point, and aspheres can have extremely high NA. Craig On Sep 29, 2016 8:33 AM, <[hidden email]> wrote:
***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/ |
|
|
In reply to this post by 0000001ed7f52e4a-dmarc-request
*****
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.
*****
In 2-photon microscopy your resolution (FWHM) should be 0.5 * wavelength / (1.414 * NA) . At 900nm and NA 1.05 this works out at 303nm. Surely if you are
just imaging beads you should be able to do better than 410nm? Guy Guy Cox, Honorary Associate Professor School of Medical Sciences Australian Centre for Microscopy and Microanalysis, Madsen, F09, University of Sydney, NSW 2006 From: Confocal Microscopy List [mailto:[hidden email]]
On Behalf Of Andreas Bruckbauer ***** 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. *****
@ Andreas: From:
[hidden email] ***** |
|
|
Michael Giacomelli |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
In reply to this post by Zdenek Svindrych-2
*****
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.
*****
A word of caution about scan lenses: they're typically designed for a narrow range of wavelengths at relatively high F/# where the effects of spherical and chromatic aberration are not pronounced. When operated outside of their design range, performance is decreased. You can get a sense of this effect just looking at the Thor catalog's scan angle vs spot size plots and comparing different wavelengths. Thorlabs is essentially the only company out there that posts (accurate) Zemax model files of their objectives on their website. Unless you are going to run them at exactly the specified wavelength and aperture, I would simulate before you build and check that everything is ok. Also worth noting, Thor's Advanced Imaging group sells scan/tube pairs that are optimized for the Ti:S range (650-1200) and for very large field numbers at very low dispersion. You used to buy individual lenses separately from the scan heads, but checking the catalog they're no longer listed. I suspect if you call and ask you can still get one at a reasonable price. These are significantly better than you will be able to construct on your own (and yes I've tried...) Mike On Thu, Sep 29, 2016 at 10:32 AM, <[hidden email]> wrote: ***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/ |
|
|
0000001ed7f52e4a-dmarc-request |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
In reply to this post by Guy Cox-2
*****
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.
*****
Guy, As I imaged at 800 and this should be treated as high NA, one would use 0.325 * lambda / (sqrt(2) * NA^0.91 and multiply with 2*sqrt(ln2) to convert from 1/e radius to fwhm (Zipfel et al 2003) which then gives 293 nm. So l calculated my own PSFs taking the vector properties and polarisation into account and got 320 nm. Convolution with the bead size of 175 nm gives roughly 365 nm (assuming Gaussian profiles). The difference between theory (365 nm) and real life (410 nm)was only 12%. It did not matter as the cells were 5 um diameter and this was all we wanted to image. Best wishes Andreas From: [hidden email] Sent: 29/09/2016 18:04 To: [hidden email] Subject: Re: Two Photon Microscope Low Signal Troubleshooting In 2-photon microscopy your resolution (FWHM) should be 0.5 * wavelength / (1.414 * NA) . At 900nm and NA 1.05 this works out at 303nm. Surely if you are
just imaging beads you should be able to do better than 410nm? Guy Guy Cox, Honorary Associate Professor School of Medical Sciences Australian Centre for Microscopy and Microanalysis, Madsen, F09, University of Sydney, NSW 2006 From: Confocal Microscopy List [mailto:[hidden email]]
On Behalf Of Andreas Bruckbauer ***** 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. *****
@ Andreas: From:
[hidden email] ***** |
|
|
*****
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.
*****
Andreas, Fair enough. When I worked with beads I used 60nm beads to side-step the convolution issue. What bugs me is that people seem to accept rather
poor resolution figures as being ‘acceptable’ when they are not. We have demonstrated 255nm FWHM resolution on collagen in SHG at 830nm (Guy Cox & Colin Sheppard, 2004. Practical limits of resolution in confocal and non-linear microscopy. Microscopy Research
& Technique, 63, 18-22). Theoretical resolution in these conditions was 210nm, and frankly I’m disappointed that we didn’t attain it. 2-photon should be comparable (actually better since it’s incoherent). These results should be attainable in everyday
use with a modicum of care. Guy
Guy Cox, Honorary Associate Professor School of Medical Sciences Australian Centre for Microscopy and Microanalysis, Madsen, F09, University of Sydney, NSW 2006 From: Confocal Microscopy List [mailto:[hidden email]]
On Behalf Of Andreas Bruckbauer ***** 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. *****
Guy, From:
[hidden email] ***** 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. *****
In 2-photon microscopy your resolution (FWHM) should be 0.5 * wavelength / (1.414 * NA) . At 900nm and NA 1.05 this works out at 303nm. Surely if you are
just imaging beads you should be able to do better than 410nm? Guy Guy Cox, Honorary Associate Professor School of Medical Sciences Australian Centre for Microscopy and Microanalysis, Madsen, F09, University of Sydney, NSW 2006 From: Confocal Microscopy List [[hidden email]]
On Behalf Of Andreas Bruckbauer ***** 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. *****
@ Andreas: From:
[hidden email] ***** |
|
|
0000001ed7f52e4a-dmarc-request |
Re: Two Photon Microscope Low Signal Troubleshooting
|
|
*****
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.
*****
Guy, I always found the small beads very dim so I was happy with the brighter ones, but you are right, one should use beads which are small compared to the expected resolution. Thanks for highlighting your SHG work, 255 is impressive, but this was with a NA 1.4 oil immersion objective. My measurement was done with a commercial instrument which I believe was quite well set up, we certainly spend a lot of time getting it right. But this raises the question of defining minimum standards for these types of microscopy. As far as I know companies do not specify resolution in their datasheets, for multiphoton or confocal microscope systems. For super-resolution systems this is different. Best wishes Andreas From: [hidden email] Sent: 29/09/2016 23:36 To: [hidden email] Subject: Re: Two Photon Microscope Low Signal Troubleshooting Andreas, Fair enough. When I worked with beads I used 60nm beads to side-step the convolution issue. What bugs me is that people seem to accept rather
poor resolution figures as being ‘acceptable’ when they are not. We have demonstrated 255nm FWHM resolution on collagen in SHG at 830nm (Guy Cox & Colin Sheppard, 2004. Practical limits of resolution in confocal and non-linear microscopy. Microscopy Research
& Technique, 63, 18-22). Theoretical resolution in these conditions was 210nm, and frankly I’m disappointed that we didn’t attain it. 2-photon should be comparable (actually better since it’s incoherent). These results should be attainable in everyday
use with a modicum of care. Guy
Guy Cox, Honorary Associate Professor School of Medical Sciences Australian Centre for Microscopy and Microanalysis, Madsen, F09, University of Sydney, NSW 2006 From: Confocal Microscopy List [mailto:[hidden email]]
On Behalf Of Andreas Bruckbauer ***** 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. *****
Guy, From:
[hidden email] ***** 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. *****
In 2-photon microscopy your resolution (FWHM) should be 0.5 * wavelength / (1.414 * NA) . At 900nm and NA 1.05 this works out at 303nm. Surely if you are
just imaging beads you should be able to do better than 410nm? Guy Guy Cox, Honorary Associate Professor School of Medical Sciences Australian Centre for Microscopy and Microanalysis, Madsen, F09, University of Sydney, NSW 2006 From: Confocal Microscopy List [[hidden email]]
On Behalf Of Andreas Bruckbauer ***** 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. *****
@ Andreas: From:
[hidden email] ***** [The entire original message is not included.]
|
| Free forum by Nabble | Edit this page |