Re: SPAD array and CLSM

[Alessandro Esposito]

*****
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 all,

  yes, I really liked that work and - of course - is a a different implementation compared to what we and others have done.

But thank Romain to remind our work, very appreciated. I am very supportive of SPAD developments, particularly those that integrate read-out and other smart things on-pixel or at least on-board. They promise to make systems more cost-effective and user-friendly while expanding their capabilities at the same time.


The main concerns with SPAD-arrays are dark counts and probability of detection. So, I have stressed the latter issue since day one of my involvement in these developments. Indeed, soon after I completed my PhD, I got in touch with David Stoppa (formerly at FBK) who then committed his group in the development of a dedicated SPAD array with high fill factor (for those days standards). Our strategy was to use a linear array and to place all photon-counting electronics on the side of the detector:

L. Pancheri and D. Stoppa, “A SPAD-based pixel linear array for high-speed time-gated fluorescence lifetime imaging,” 2009 P ESSCIRC, 429–432 (2009).

Then, we used the system for time-resolved spectroscopy collaboratively with Clemens Kaminsky's lab. It took several year for me to be able to integrate it in a microscope, with FBK developing  a firmware that would preserve as many photon counts as possible and we arrived to the manuscript you have mentioned.

A brilliant strategy to overcome the issue related to low probability of detection was implemented by Simon Ameer-Beg's group (as you pointed out). Rather than worrying about fill factor, they exploited this in a multi-beam confocal FLIM system. I always mentioned that I regret I did not have that idea!

Anyway, technologies are getting better and better and fill factors are going up. I should highlight the massive contribution to the field from Robert Henderson's group (Edinburgh) and Edoardo Charbon's group (EPFL) both developping - among other things - wonderful smart pixels for microscopy applications.

I am sure I am missing out many other valuable contributors to the field (here I focused on my current knowledge for time-resolved arrays for confocal microscopy and with a perspective on increasing fill factor), but I thought to give a contribution as I was mentioned. All this to arrive to my next statements :)


The issue related to limited PDE and fill-factors, particularly in 2D arrays with embedded in-pixel electronics, is gradually disappearing and, therefore, I agree with Giovanni in that we might be at a turning-point in the field. At last, microlens arrays seems to yield but also back illuminated sensors where the electronincs do not steal space to the photo sensitive areas, bonded chips and so on will make these technologies super-effective.


Personally, while I see 'wide-field' technologies (SIM, light-sheet, camera-based super-res) making point-scanning systems a tiny bit less important nowadays, I would argue that multi-parametric imaging of various type, including beautiful examples such as Giuseppe's, can be enabled by detector technologies (SPAD is one prominent examples) only - or at least more practically - in point scanning systems. Hence, I think this is the direction that confocal microscopes will keep going towards.