Rapid acquisition, analysis, and display of fluorescence lifetime-resolved images for real-time applications

Abstract

Fluorescence lifetime-resolved imaging (FLI) is a relatively new technique of fluorescence imaging whereby the spatial distribution of fluorescence decay times can be determined directly at every pixel of an image simultaneously. The fluorescence decay times of many chromophores can act as sensitive gauges of their molecular environments. By employing measurement techniques that are quantitatively related to the radiative dynamics of the dye molecules (in the nanosecond time range), additional physical parameters are available for discerning different fluorophores with disparate lifetimes, or for characterizing a single fluorophore in different surroundings. Many physical processes such as molecular aggregation, binding of dyes to macromolecular species, inclusion of chromophores in specific cellular organwelles, fluorescence resonance energy transfer, and dynamic quenching determine the excited-state lifetime of a fluorophore. The FLI technique provides a way to measure these processes directly at 103–106 pixels in an image. In addition, if image domains differ with respect to the mean fluorescence lifetime, FLI can be used to improve the contrast of a fluorescence image. By measuring the fluorescence lifetime one can determine whether fluorescence intensity differences from different locations in an image can be attributed to differences in dye concentration or whether physical spectroscopic effects such as local differences in the rate of dynamic quenching are responsible. All the above applications provide new possibilities for biology and medical diagnostics. However the speed of data acquisition and analysis in current FLI instrumentation is limited in general to several minutes; for real-time applications (in order to follow rapid changes of microscopic samples or make in vivo endoscopic medical diagnosis) the present instruments are too slow. We present here a FLI apparatus that is capable of acquiring, processing, and displaying fluorescence lifetime-resolved images in quasi-real time. We also present rapid algorithms for analyzing the data in real time.