Facebook Twitter Instagram YouTube

Optical and computational super-resolution methods for dense scenes in fluorescence microscopy

Event Details

Event Dates: 

Monday, November 18, 2013 - 4:00pm

Seminar Location: 

  • Engineering Building Room ECCR 105

Speaker Name(s): 

Tony Barsic

Speaker Affiliation(s): 

University of Colorado Boulder
Seminar Type/Subject

Scientific Seminar Type: 

  • COSI Seminar

Event Details & Abstract: 

Super-resolution localization microscopy is an imaging paradigm in which the images of individual emitters (e.g. fluorescent molecules, beads, or quantum dots) are fit to a model of the system point spread function (PSF) so that their location can be estimated with high precision. A super-resolution image is then built up in a pointillistic fashion by localizing many molecules. Experimental techniques exist to control the density of activated molecules in a given frame, so that the majority of emitters can be in an “off” state. These procedures are used to ensure the PSFs are not overlapping, thus enabling the use of simple fitting schemes for location estimation. The disadvantage of such time multiplexing is long data collection times. Through a combination of optical and computational techniques, we have developed methods to enable accurate localization of dense scenes. One technique in particular can localize point sources in three dimensions, and enables a 10-fold increase in the allowable label density, potentially reducing imaging times accordingly.