Maximum permissible exposure in confocal or multiphoton microscopy

[Dan Fu-2]

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Dear all,

Recently I was calculating the maximum permissible exposure (MPE) power
level for confocal or multiphoton microscopy for tissue imaging. It seems to
me that currently the optical power used by most confocal or multiphoton
studies in tissue imaging far exceeds the MPE limit. Here is how I
calculated it and please tell me if I am doing something wrong. The
literature I have found so far do not talk about tight focusing.

According to the ANSI-Z136.1 standard, in the NIR range with an exposure
time of 10^-7 to 10 second, the MPE level is 0.56*t^0.25, t being the pixel
dwelling time. Considering a typical pixel dwelling time at 10 microsecond,
the MPE level is 31 mJ/cm-2. Assume we have a focal spot size of 1 micron,
the average power should be 0.31 nJ or 0.031 mW. That seems like an awfully
small number considering a few mW (even tens of mW) is routinely used in
human skin imaging.

If some one has worked on this before, please advise me what is wrong here.
Thank you very much for your input.

Dan Fu

[George McNamara]

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hi Dan,

Permissible to who?

Once you reach fluorophore saturation there is no need to go higher
for confocal (I believe SHG and THG doe not saturate). So, if laser
power X and X+10% (i.e. 90 and 100% AOTF setting) give equal
brightness (assuming PMT gain not trivially saturated), then X is saturated.

If your 543 nm laser routinely achieves fluorophore saturation, let's trade.

If a 1k x 1k scan takes 1 second, pixel dwell time is 1 usecond or
less (ignoring spot size wrt Nyquist), so your 10 us is an over-estimate.

George
p.s. of course if you want to use saturation to achieve
ultramicroscopy resolutions (Gustafsson, et al for example), then
X+20% might be better than X.


At 11:33 AM 7/7/2008, you wrote:

>Search the CONFOCAL archive at
>http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
>Dear all,
>
>Recently I was calculating the maximum permissible exposure (MPE) power
>level for confocal or multiphoton microscopy for tissue imaging. It seems to
>me that currently the optical power used by most confocal or multiphoton
>studies in tissue imaging far exceeds the MPE limit. Here is how I
>calculated it and please tell me if I am doing something wrong. The
>literature I have found so far do not talk about tight focusing.
>
>According to the ANSI-Z136.1 standard, in the NIR range with an exposure
>time of 10^-7 to 10 second, the MPE level is 0.56*t^0.25, t being the pixel
>dwelling time. Considering a typical pixel dwelling time at 10 microsecond,
>the MPE level is 31 mJ/cm-2. Assume we have a focal spot size of 1 micron,
>the average power should be 0.31 nJ or 0.031 mW. That seems like an awfully
>small number considering a few mW (even tens of mW) is routinely used in
>human skin imaging.
>
>If some one has worked on this before, please advise me what is wrong here.
>Thank you very much for your input.
>
>Dan Fu

George McNamara, Ph.D.
University of Miami, Miller School of Medicine
Image Core
Miami, FL 33010
[hidden email]
[hidden email]
305-243-8436 office
http://home.earthlink.net/~pubspectra/
http://home.earthlink.net/~geomcnamara/
http://www.sylvester.org/research/SR_lab_analytical.asp?ana=desc 
(Analytical Imaging Core Facility)

[Julian Smith III]

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Dan--I can only help on the front end of the
calculation--with our FV1000, we typically run a
pixel dwell of 4µs; on the new Zeiss 710 (same
specimens), it was more like 1µs, so 10µs seems
high.
Julian

>Search the CONFOCAL archive at
>http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
>Dear all,
>
>Recently I was calculating the maximum permissible exposure (MPE) power
>level for confocal or multiphoton microscopy for tissue imaging. It seems to
>me that currently the optical power used by most confocal or multiphoton
>studies in tissue imaging far exceeds the MPE limit. Here is how I
>calculated it and please tell me if I am doing something wrong. The
>literature I have found so far do not talk about tight focusing.
>
>According to the ANSI-Z136.1 standard, in the NIR range with an exposure
>time of 10^-7 to 10 second, the MPE level is 0.56*t^0.25, t being the pixel
>dwelling time. Considering a typical pixel dwelling time at 10 microsecond,
>the MPE level is 31 mJ/cm-2. Assume we have a focal spot size of 1 micron,
>the average power should be 0.31 nJ or 0.031 mW. That seems like an awfully
>small number considering a few mW (even tens of mW) is routinely used in
>human skin imaging.
>
>If some one has worked on this before, please advise me what is wrong here.
>Thank you very much for your input.
>
>Dan Fu

--
Julian P.S. Smith III
Director, Winthrop Microscopy Facility
Dept. of Biology
Winthrop University
520 Cherry Rd.
Rock Hill, SC  29733

803-323-2111 x6427 (vox)
803-323-3448 (fax)
803-524-2347 (cell)

[Dan Fu-2]

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hi George,

Sorry I wasn't being clear in the original post. Maximum Permissible
Exposure (MPE) is a term used in the American National Standard for Safe Use
of Lasers (ANSI Z136.1). When confocal or multiphoton microscopy is used in
human skin imaging, I assume the standard for skin exposure should also be
followed.

In both reflectance Confocal laser scanning microscopy and multiphoton
microscopy of human skin (in vivo), the optical power used are usually more
than 10 mW with very tight focusing condition. I am not sure about the
saturation problem you mentioned, but I would think that the only reason to
use higher power is to increase the signal to noise ratio. The pixel time
can be reduced to 1 microsecond, but even with that number, the calculated
MPE is 0.17 mW. This means that the routine skin imaging conditon of
confocal or multiphoton is way above the national safety standard.

I am interested in knowing whether there is anything wrong in my calculation
or if the national standard is inappropriate in this case.

For reference in reflectance confocal or multiphoton, see
1. Langley, R. G. B.; Rajadhyaksha, M. et. al., Journal of the American
Academy of Dermatology 2001, 45, (3), 365-376.
2. Konig, K.; Riemann, I. et. al.,Journal of Biomedical Optics 2003, 8, (3),
432-439.

Regards,
Dan

[Julio Vazquez]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hi Dan, 

First of all, it would help if you clearly specified in what units you are measuring the MPE (I suppose in energy units). I don't know what the MPE is supposed to measure or how it was established (well, I guess I could read the papers), and it probably matters whether the units are power units or energy units. However, it may also matter whether you are delivering a high power over a short time, or a low power over a long time... I'd rather expose my skin to daylight for 1 day, than to a 10 Watt laser beam for 1 second, even though the dose administered by the laser may be lower...

In any event, 10 mW corresponds to 10 mJ/sec or 10 nJ/microsecond. So if you are using 10 mW of power, you are delivering 10 nanoJoules of energy to your sample for every microsecond of exposure. 

This being said, you need to take into account the efficiency of your confocal system. Typically, the amount of power at the sample will be much less than the nominal power delivered by the laser. On our system, it runs around 20-35%, depending on the specific laser, objective, etc... So you probably need to do your calculations with the power you measure at the sample with a power meter, and not the power you think you are delivering based on you laser nominal output and your AOTF settings.

MPE aside, I would consider two factors:

1. are you using more laser power than needed to excite your fluorophore (i.e. saturating/photobleaching it faster than necessary). In that case, I would recommend a practical approach such as mentioned by George: do a response curve (measured fluorescence intensity versus laser power); you probably will see a linear response followed by saturation. Try to stay in the (early part of) the linear curve (unless saturation is desired)

2. are you damaging (biologically) your sample, so that results may no longer be reliable. In that case, you may want to have an assay to evaluate the degree of happiness of your sample (I don't know what that would be for a piece of skin). For example, I have used cell division, or mitochondria depolarization (revealed by leaking out of mitochondrial reporter dyes) as viability assays, and typically I would use laser powers less than 50% (ideally less than 25%) the amount that gives me a visible effect on cells. That's a rule of thumb that works for me...

From my experience, if you are well below saturation levels for a typical fluorescent dye, you are probably safe in terms of the biology too, but that may depend on many factors. When we use a 488nm Argon laser, we try to stay definitely below 0.2 mW at the sample (10% at the AOTF with a 10mW laser), and ideally below 0.1 mW. As George also says, it's hard to saturate a dye with a 543 HeNe laser even at full power (1 mW nominal, probably around 0.2-0.3 mW at the sample), although we also use less than full power whenever we can. With 2 photon excitation, I haven't have as much experience, but I try to keep it below maybe 20mW at the sample

--

Julio Vazquez

Fred Hutchinson Cancer Research Center

1100 Fairview Ave. N.,  mailstop DE-512

Seattle, WA 98109-1024

[hidden email]

http://www.fhcrc.org/

==

On Jul 7, 2008, at 8:33 AM, Dan Fu wrote:

Search the CONFOCAL archive at

http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Dear all,

Recently I was calculating the maximum permissible exposure (MPE) power

level for confocal or multiphoton microscopy for tissue imaging. It seems to

me that currently the optical power used by most confocal or multiphoton

studies in tissue imaging far exceeds the MPE limit. Here is how I

calculated it and please tell me if I am doing something wrong. The

literature I have found so far do not talk about tight focusing.

According to the ANSI-Z136.1 standard, in the NIR range with an exposure

time of 10^-7 to 10 second, the MPE level is 0.56*t^0.25, t being the pixel

dwelling time. Considering a typical pixel dwelling time at 10 microsecond,

the MPE level is 31 mJ/cm-2. Assume we have a focal spot size of 1 micron,

the average power should be 0.31 nJ or 0.031 mW. That seems like an awfully

small number considering a few mW (even tens of mW) is routinely used in

human skin imaging.

If some one has worked on this before, please advise me what is wrong here.

Thank you very much for your input.

Dan Fu

[Dan Fu-2]

Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hi, Julio. Thank you for your helpfup input. I am not working on measuring
the MPE level or anything like that. I am interested in learning for human
skin imaging application, what is a safe power level? There are some
experimental studies on that, but I was wondering if the ANSI standard also
applies here since it has specific guideline on eye and skin exposure limits.

I think my question can be simplified like this: for a CW laser at 800nm,
how much optical power can be deposited into a 0.25 um2 area in 1 us without
raising damage concerns? The units should be energy unit or converted to
average power by dividing the time duration. My calculation shows that

So the question is more about safety concern rather than saturating the
fluorophores. And the sample here is animal model or human, not cell
samples. Julio has very good suggestions about how to monitor the sample
damage. For now, I would just like to have an idea whether the existing
standard is applicable and if yes, how do I calculate the maximum power limit.

I really appreciate any thoughts on this. If you feel this topic is not
appropriate for this forum, please reply me off the list. Sorry to bother
those who are not interested.

Best Regards,
Dan