Brain Function: Novel Technologies Driving Novel Understanding - Bioinspired Approaches for Human-Centric Technologies - page 303

Fig. 10.4 Vignetting super-resolution approaches—targeted and stochastic readout versus con-
ventional imaging. Fluorescent molecule distribution as it appears in conventional imaging,
diffraction limited, and as it can be spatially super-resolved (Credit: G. Vicidomini)

(Fig. 10.4 ). Stochastic readout is related to individual molecule localization
methods like PALM (Betzig et al. 2006 ), FPALM (Hess et al. 2006 ), STORM
(Rust et al. 2006 ), GSDIM (Folling et al. 2008 ), or 3D IML-SPIM (Cella Zanacchi
et al. 2011 ). These approaches create a sparse distribution of point-like emitters,
single molecules, and localize the emitters with precision of ~20 nm in the x - y
dimensions and 50 nm along the z dimension. Photoactivatable or photoswitchable
molecules are used, and the incident light intensity is reduced to lower the proba-
bility of the photoactivation process, thus creating a sparse distribution of fluores-
cently activated single molecules. Sparse refers to the fact that activated individual
molecules are, with high probability, separated by distances larger than the diffrac-
tion limit. Data collection is wide field, requiring the collection of hundreds to
thousands of frames, and the spatial localization is typically realized off-line by
statistical analysis (Thompson et al. 2002 ; Mortensen et al. 2010 ).
As long as the number of photons collected for each emitter ( N ) is sufficient, the
fluorophore position can be determined with ten times higher precision than with
conventional imaging.
The general relationship is given by

p N

s ¼ s 0 =

where s is the localization precision, s 0 is the size of the diffraction-limited