Stochastic Optical Fluctuation Imaging (SOFI) is a superresolution
fluorescence microscopy technique which allows to enhance the
spatial resolution of an image by evaluating the temporal fluctuations of
blinking fluorescent emitters. SOFI is not based on the identification and
localization of single molecules such as in the widely used Photoactivation
Localization Microsopy (PALM) or Stochastic Optical Reconstruction
Microscopy (STORM), but computes a superresolved image via temporal
cumulants from a recorded movie. A technical challenge hereby is that,
when directly applying the SOFI algorithm to a movie of raw images,
the pixel size of the final SOFI image is the same as that of the original
images, which becomes problematic when the final SOFI resolution is much
smaller than this value. In the past, sophisticated cross-correlation schemes
have been used for tackling this problem. Here, we present an alternative,
exact, straightforward, and simple solution using an interpolation scheme
based on Fourier transforms. We exemplify the method on simulated and
experimental data.
|