Theory of Structured illumination microscopy

[u0362565]

Hi all,

My question concerns the theory of how super resolution microscopy causes a frequency shift to allow information normally outside of the observable region of frequency space to be captured by the objective lens.

I have read various papers on this including those by Mats Gustafsson which describe the theory but I'm still having a hard time trying to visualise what's actually going on.

I'd like to be able to describe to people who know nothing about optics how this method works and the angle i'd like to take is that its based on the interference of light. But for my own understanding i'd like to know why moire patterns form when the excitation light is "structured" i.e. contains a high spatial frequency illumination pattern.

When trying to visualise this, a striped pattern would be seen on the sample as regular repeating lines. In the light bands excitation light is exciting the sample but where there is a dark band there is no excitation?

But what i don't really understand is where is the interference coming from? Why is it not simply that there is emission from the light areas of the grid but none from the dark areas.

Its easy to explain beat frequencies in terms of sound and as i understand it beat frequencies in sound and moire patterns are the same thing? When 2 sounds of similar frequency interfere the result is the difference between the 2 frequencies. When thinking of light and structured illumination though its not such a simple case. As the frequencies we're referring to are not the frequencies of light as we traditionally think of it i.e. wavelengths per second, but the spatial frequency of the illumination pattern?

Everything i read talks about the interaction of a high spatial frequency introduced by the incident excitation light and the high spatial frequencies inherent in the distribution density of fluorescence, but to me this isn't a good enough explanation for what's going on. Can anyone explain this in a straightforward way by describing what's actually happening between the incident light and the emitted light to cause moire patterns and thus a frequency shift? Perhaps by replicating the idea of how this occurs with sound.

Thanks for your help!

Matt

 

[Danger]

I didn't know that such a thing exists. Thanks for posting this. I'm interested in seeing the results.

 

[Andy Resnick]

For a relatively painless introduction, try:

http://zeiss-campus.magnet.fsu.edu/tutorials/superresolution/hrsim/index.html
http://cbst.ucdavis.edu/research/bioimaging/sim

Gustafsson's paper is here:

http://www.pnas.org/content/102/37/13081.full.pdf+html

 

[u0362565]

I like gustafsson's explanation and this paper has the clearest description I've seen. I still feel though that it doesn't explicitly explain why the moire patterns form. Simply saying frequencies mix to produce the difference frequency doesn't seem like enough or should i just accept this as is.