*****
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.
*****
OMX is old tech (for example, the deconvolution is from around 1995).
OMX used multiple PC's (one per camera if I recall correctly).
Multiple cameras ... Can optimize exposure times and illumination
intensity to get the most and best data. GPU enables instant
gratification spatial deconvolution (and hopefully "joint" processing).
Fluorescence imaging was at 7plex (albeit inefficiently) in 2000,
http://jhc.sagepub.com/content/48/5/653.long
has not progressed nearly as much as flow cytometry.
On 3/19/2016 4:03 PM, Andreas Bruckbauer wrote:
> A good number of cameras would be eight sCMOS 82% QE cameras (i.e. USB3).
>
> I guess this would be too much data to transfer. The OMX had four
> computers for four cameras, not sure if this is still necessary, but
> eight fast cameras with lots of pixels sounds to me difficult to
> handle. Maybe two quad-splitter with a camera splitter combined?
> Best wishes
>
> Andreas
> ------------------------------------------------------------------------
> From: George McNamara <mailto:[log in to unmask]>
> Sent: 19/03/2016 20:15
> To: [log in to unmask]
> <mailto:[log in to unmask]>
> Subject: Re: Service contracts
>
> *****
> 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 Avi, -- and I am copying this to the listserv,
>
> Back when I was in Miami, one of the Leica confocal applications
> scientists (I forget his name - sorry!) showed me side by side
> comparison of resonant scanner (RS) vs standard mode for as identical
> acquisition settings as possible, on a couple of specimens. The bottom
> line was that for the same total dwell time, RS gave as good (or better)
> data, with less photobleaching of those particular specimens. My
> recollection is that very slow dwell times in standard mode led to a lot
> more photobleaching.
> I encourage you / your lab / your users (maybe all late first year grad
> students should do this and other exercises on shared instrumentation!)
> to compare:
> standard scanner:
> * 80 Hz (if slowest dwell time)
> * 800 Hz (I forget whether this is Leica SP5 default speed)
> * max standard scanner speed (for say 512x512 pixel image)
> RS = 8000 Hz
> * no averaging
> * 10 frame averaging (effectively 800 Hz)
> * 100 frame averaging (effectively 80 Hz
> Personally, I hate averaging on a confocal ... long ago I complained on
> the Confocal Listserv that the confocal manufacturers should come up
> with "more sophisticated" algorithms for denoising confocal PMT data ...
> as the simplest improvement, I suggested odd number of acquisitions, and
> taking the median for each pixel (this way would be a real pixel value,
> not average of two intensity reads). For example, PMT at high gain might
> output, 101, 1, 1, 1, 1, average = 21, median = 1.
> Note: RS mode is even more fun with very few rows, single line, so 8000
> linescans per second on Leica SP5.
>
> One of my favorite papers (except for their acronym [it would work on
> non-FRET data] and slow speed in 2008)
> http://www.ncbi.nlm.nih.gov/pubmed/?term=hoppe+swanson+2008
> PDF available at
> http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2426648/
> 61 lines of Matlab code (though needs a commercial library):
> The computer simulations and algorithm development were carried out in
> using MATLAB 7.3 (The Mathworks, Natick, MA) and the DIPImage toolbox
> for MATLAB (http://www.diplib.org/, Quantitative Imaging Group, Delft
> University of Technology, Netherlands) in the Linux operating system
> (OpenSUSE 10.0, Novell, Waltham, MA). The computer was a custom-built
> dual-Athlon (AMD) machine with 5 GB RAM. The MATLAB function for 3DFSR
> can be found athttp://sitemaker.umich.edu/4dimagingcenter/3dfsr.
> (oops, that UMich web site ended - email the authors for the .M code).
>
> -- what I refer to as "joint spatial deconvolution and spectral
> unmixing" (JSDSU or JSDSUN) ... results in a 10x improvement in signal
> to noise ratio. Spatial deconvolution is now "instant gratification"
> with GPU(s):
> www.microvolution.com (yes, that's my image on their home page - no
> financial interest) and presumably (I've not tested it)
> SVI Huygens w/ GPU ... https://svi.nl/HuygensGPU
>
> NVidia top of the line GPU graphics card, GeForce GTC Titan-X, is around
> $1000 (and drives an HD 4K monitor). By end of 2016 their next gen card
> will be same $1000, but about 10x faster (more cores, somewhat higher
> clock rate) and more RAM ... and some PC's could take 2 or 3 of these.
>
> I hope both Microvolution and Huygens implemenet joint processing.
>
> I would also like to see the microscope vendors implement many-cameras
> to use the full spectral range of fluorescence, full field of view,
> without moving parts. Also as compact as possible, and simple C-mount to
> the microscope.
>
> A good number of cameras would be eight sCMOS 82% QE cameras (i.e.
> USB3). The usable fluorescence emission range is about 370-850 nm, so
> 480 nm range divided by 8 cameras is about 60 nm per emission band.
> Probably narrower in the cyan-green-yellow bands, and much wider in
> the NIR.
>
> One could today purchase two Cairn Research MultiCam LS's
> https://www.cairn-research.co.uk/product/multicam-ls/
> replace one of the eyepiece/camera mirrors with an appropriate dichroic,
> and go 8plex.
>
> S.J. Morris published in the early 1990's on 4 ICCDs for simultaneous
> Ca++ and pH ratiometric imaging (Nikon splitter on the Diaphot, followed
> by splitters on each of those ports).
>
> Jeff Price (Sanford-Burnham) and Sally Ward & Raimund Ober (long time
> UTSW, now at Texas A&M) have published on multi-focal setups. If
> academics can do it, how about vendors stepping up?
>
>
> George
> p.s. one TITAN X price is $1251,
> http://www.amazon.com/GeForce-Graphics-384-Bit-Express-GTXTITANX-12GD5/dp/B00V7C9N26
> GeForce GTX TITAN X Graphics Card, 12GB GDDR5 384-Bit, PCI Express 3.0
> HDCP Ready SLI Support Video Card (GTXTITANX-12GD5)
> Academic pricing might be better.
>
> p.p.s. Using a coverglass with a sparse distribution of
> ThermoFisher/.../Molecular Probes TetraSpeck beads (40 or 100 nm
> diameter), and/or MM2 nanoparticles (the latter also provides O2 data,
> first emission band is similar to BV421, O2 sensitive band is narrow at
> 650 nm ... ok, so maybe 12 cameras, also Tb and Eu lanthanides
> emissions, and might as well have a full time dedicated transmitted
> light band, such as way out in NIR, re: Dodt's IR-VEC-DIC, though I
> would just go brightfield + deconvolution and/or QPm),
> http://luxcel.com/MitoImage%C2%AE+MM2+Description
> http://ibidi.com/fileadmin/products/cells_reagents/R_741XX_NanO2_MM2/IN_74161_MM2.pdf
> If the NIRvana Sciences chlroins/chlorphyll based dyes work well (their
> academic cofounders, Jonathan Lindsey et al, are now at QY 0.3 for
> some), these narrow emission peaks,
> http://nirvanasciences.com/?page_id=3088
> especially if could be tandem'd on the Brilliant's (BUV, BV, BB ...
> maybe some day BG?) could enable imaging with as many plex (or more)
> than the flow folks.
>
>
> On 3/17/2016 1:02 AM, Avi Jacob wrote:
> > Hi George!
> >
> > So, about the RS. I don't use mine that often, but it seems you
> > recommend using it when doing large Zs?
> > It does give a grainier image, even with averaging. And if one used a
> > LOT of averaging, you lost the speed.
> > I use it mostly when doing Live.
> >
> > Avi
> >
> >
> > --
> > Avi Jacob, Ph.D.
> > Head of Light Microscopy
> > The Mina & Everard Goodman Faculty of Life Sciences
> > Bar-Ilan University, Ramat-Gan 5290002, Israel
> > Cell: 052-5802544 (call here first), Desk: 972-3-5317647
> > http://tinyurl.com/BIU-Microscopy
> >
> >
> >
> >
> > On Thu, Mar 17, 2016 at 7:21 AM, George McNamara 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.
> > *****
> >
> > Hi Andy,
> >
> > //
> >
> > My biggest contribution to microscopy at the U was when I urged
> > one of the MP/SP5 users who did 'islets in the eye' confocal
> > imaging (and did so with modest number of channels and pixel
> > count, whatever size Z-series), to try using the Leica SP5
> > resonant scanner instead of the default 'standard' (aka slow)
> > scanner. That user waved me off. Several days later I asked the
> > faculty member in charge of that project about RS. The faculty
> > member replied that the acquisitions were now so fast that the
> > user did not have the opportunity to take breaks during the
> > Z-series. They now have 10 papers together. The faculty member
> > was - an is - an assistant professor at the U. The user is now an
> > assistant professor at the U.
> >
> > George
> > p.s. "newer, better" reminded me of this article,
> > D. Ward 2012Faster, Better, Cheaper: Why Not Pick All Three?
> > National Defense Magazine.
> >
> http://www.nationaldefensemagazine.org/archive/2012/April/Pages/Faster,Better,CheaperWhyNotPickAllThree.aspx
> >
> > On 3/16/2016 1:11 PM, Andrew York 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.
> > *****
> >
> > Hello listserv, I'm interested in your opinions regarding
> > service
> > contracts on high-end commercial microscopes in multi-user
> > core facilities.
> > What has your experience been with service contracts? Any
> > mistakes or
> > regrets? Any advice on negotiating pricing, or if/when to drop
> > contracts?
> >
> > I've read through some old posts on this topic, but I bet
> > there's more
> > useful knowledge lurking out there, and I bet I get some juicy
> > off-list
> > replies like usual.
> >
> > Thanks, as always.
> > -Andy
> >
> >
> >
> > --
> >
> >
> >
> > George McNamara, Ph.D.
> > Single Cells Analyst, T-Cell Therapy Lab (Cooper Lab)
> > University of Texas M.D. Anderson Cancer Center
> > Houston, TX 77054
> > Tattletales http://works.bepress.com/gmcnamara/42
> > http://works.bepress.com/gmcnamara/75
> > https://www.linkedin.com/in/georgemcnamara
> >
> >
--
George McNamara, Ph.D.
Single Cells Analyst, T-Cell Therapy Lab (Cooper Lab)
University of Texas M.D. Anderson Cancer Center
Houston, TX 77054
Tattletales http://works.bepress.com/gmcnamara/42
http://works.bepress.com/gmcnamara/75
https://www.linkedin.com/in/georgemcnamara
|
