Re: [QUAR] Re: widefield getting better images than spinning disk

Posted by Andrea Latini on
URL: http://confocal-microscopy-list.275.s1.nabble.com/widefield-getting-better-images-than-spinning-disk-tp7585177p7585212.html

*****
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.
*****

With full respect for the below named Vendors please allow me to specify
that my below reported measurements were not carried out at our labs nor
confirmed by us (we did confirm only our system final results, of course).
Enduser carried them out in specific experimental conditions that were not
under our control.
Reason for posting them was in trying to answer Zdenek's question about
'what is useful piece of information for the customer' in using such
measurement protocols: is some specific case they 'could' be used for system
comparison and performance evaluation under experimental condition of own
interest.

Hence, they must not be intended as different Vendor's Confocal System
'overall and general' performance comparison.

regards.


Andrea Latini
President
CrestOptics Srl


On Mon, 23 May 2016 15:13:13 -0500, Andrea Latini <[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.
>*****
>
>dear Zdenek,
>I guess this comparison, we've got from enduser, would be interesting for
>others.
>using same protocol.
>
>"..
>- Throughput confocal performance ratio (CF/WF %), taking into account both
>excitation and collection efficiency:
>
>o Visitech 1% efficiency
>o Yoko CSU X1  2.8% efficiency
>o Crest X-Light 5.5%-6.3% efficiency
>
>Note: we have been using 200nm beads, 100X 1,49 obj and SpectraX LED system
>(not even lasers), Evolve CCD
>.."
>
>regards.
>
>Andrea Latini
>President
>CrestOptics Srl
>
>
>
>On Mon, 23 May 2016 21:21:14 +0200, Zdenek Svindrych <[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.
>>*****
>>
>>Dear Andrea,
>quite off the original topic, but, whatever...
>
>As already mentioned, it's very important to separate the overall efficiency
>into excitation efficiency and collection efficiency.
>
>5% excitation efficiency would be tolerable, e.g. with a spinning disk
>without microlenses...
>
>But in detection path "every photon counts", 80% efficiency should be
>possible with > 1AU pinholes.
>
>What's missing in the protocol is "measure the illumination intensity in the
>BFP (or image plane) of the objective in both cases and do the math"...
>
>And what would be the useful piece of information for the customer? Perhaps
>something like "increase tour illumination power 27.35 times (don't quote me
>here :-) to get the same average illumination intensity if you want to
>compare spinning disk to widefield..."
>
>Best, zdenek
>
>
>
>
>
>
>---------- Původní zpráva ----------
>Od: Andrea Latini <[hidden email]>
>Komu: [hidden email]
>Datum: 23. 5. 2016 14:40:42
>Předmět: Re: [QUAR] Re: widefield getting better images than spinning disk
>
>"*****
>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.
>*****
>
>dear Nico,
>the spinning disk itself acts as an optical filter on thick samples.
>Spinning Disk fill factor means 'Open Area Ratio' - holes area / blocked
>area.
>by inserting the disk within the optical path the theoretical throughput
>collection value would be 8.2% with respect to widefield (both excitation
>and emission pass through spinning disk of course), on 4 um beads, GFP
>excitation channel.
>
>the value 5-6% on 4um beads (and other values I'm reporting), has been
>measured by using "Measuring and interpreting point spread functions to
>determine confocal microscope resolution and ensure quality control Richard
>W Cole,Tushare Jinadasa,& Claire M Brown; Nature Protocols, Volume 6, Pages:
>1929–1941(2011)" method.
>
>briefly (Macro is running in NIS Element SW):
>"
>
>1.) Sample: subresolution beads prepared according to „Measuring and
>interpreting point spread functions to determine confocal microscope
>resolution and ensure quality control
>Richard W Cole,Tushare Jinadasa,& Claire M Brown; Nature Protocols, Volume
>6, Pages: 1929–1941(2011)
>2.) Technical setup of Microscope:
>a. The microscope is set up for a typical point spread function
>acquisition to characterize the resolution/imaging quality using a high-end
>100x oil objective.
>b. A regular fluorescence light source is used for all acquisitions in
>combination with a filter cube. This is essential and the bottom principle
>of the test. The same illumination with the same sample is used to compare
>wide field intensity with the confocal intensity of bead images. This test
>assesses only the emission light transmission of the spinning disc device.
>i. Option one preferred: left and right port of the microscope is used so
>that one port holds the spinning disc device with camera and the other port
>an identical second camera in simple wide field mode.
>ii. Alternatively one can run the acquisition of test data first with the
>spinning disc +camera and then remove the scan head and use the same camera
>alone on the same port where the spinning disc scan head was mounted.
>iii. Toggle between wide field mode and spinning disc mode if the to be
>tested scan head provides such an option and (important!) if the wide field
>mode does not have more than a single mirror in the light path (otherwise
>the transmission loss of the additional optical elements in the wide field
>mode of the scan head results in a lower intensity measurements . This in
>turn would artificially “improve� the result so that one would think the
>ratio wide field intensity / confocal intensity is higher and therefore
>pretending to be the better instrument)
>
>3.) Procedure
>Time required: about 1 -2 hours
>Overview of work flow: acquisition of image z-stacks of beads at 3 different
>exposure times in each mode (confocal and wide field) while illumination
>remains constant.
>Prepare collapsed stack images (max projection)
>rolling ball background reduction
>find beads and set ROIs
>calculate mean or median intensity (in our hands it does not make a
>difference)
>plot intenstity vs. exposure time for both modes
>determine slope
>ratio of slopes = result
>
>In detail:
>a.) Set up beads in focus with the scanning device. The signal should be
>chosen at an intensity (illumination light intensity ) , so that with the
>same illumination wide field images are possible (sufficiently short shutter
>time should be available)
>b.) Stacks of images are taken at 3 different exposure times of different
>view fields to avoid the issue of bleaching,
>c.) Repeat the steps a) and b) in wide field mode
>d.) Run macro in ImageJ (enclosed) to analyze bead intensity
>
>Find different ROIs in each maximum projection :
>
>run("Z Project...", "projection=[Max Intensity]");
>run("Subtract Background...", "rolling=5");
>run("Threshold...");
>title = "WaitForUserDemo";
>msg = "If necessary, use the \"Threshold\" tool to\nadjust the threshold,
>then click \"OK\".";
>waitForUser(title, msg);
>getThreshold(lower, upper)
>run("Analyze Particles...", "size=1-4 exclude summarize");
>
>
>or: you run the macro above on the stack with the lowest exposure time
>and then – to keep the ROIs the same for all stacks- yu run the macro below
>on the remaining stacks
>
>run("Z Project...", "projection=[Max Intensity]");
>run("Subtract Background...", "rolling=5");
>roiManager("measure")
>
>
>e.) Prepare plots of the so calculated intensities vs the exposure time
>f.) Calculate slope
>g.) Ratio the slope.
>
>"
>
>
>regards.
>
>Andrea
>
>Andrea Latini
>President
>CrestOptics Srl
>
>
>On Mon, 23 May 2016 08:42:07 -0700, Nico Stuurman <[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.
>>*****
>>
>>Dear Andrea,
>>
>>> collection efficiency CF/WF is 5% to 6% in our systems with 70um pinholes
>(fill factor 8.2%).
>>
>>Could you please explain what that number means, or better, how it is
>>measured? The way it is written, I have the impression that when a very
>>thin film of fluorescent material is excited by the same amount of light
>>you get 5-6% intensity in confocal mode and 100% in wide-field mode.
>>Obviously, that would be pretty bad (and according to the discussion on
>>this list, you should not lose more than a few percent of the
>>in-focus-signal through the pinholes), so I assume that you mean
>>something else.
>>
>>Best,
>>
>>Nico"