Kevin McHale | PubMed Health

PubMed2007

Feedback localization of freely diffusing fluorescent particles near the optical shot-noise limit.

We report near-optimal tracking of freely diffusing fluorescent particles in a quasi-two-dimensional geometry via photon counting and real-time feedback. We present a quantitative statistical model of our feedback network and find excellent agreement with the experiment. We monitor the motion of a single fluorescent particle with a sensitivity of 15 nm/sqrt Hz while collecting fewer than 5000 fluorescence photons/s. Fluorescent microspheres (diffusion coefficient 1.3 microm2/s) are tracked with a root-mean-square tracking error of 170 nm, within a factor of 2 of the theoretical limit set by photon counting shot noise.

Journal Article↗

PubMed2004

Bayesian estimation for species identification in single-molecule fluorescence microscopy.

In this article we describe a recursive Bayesian estimator for the identification of diffusing fluorophores using photon arrival-time data from a single spectral channel. We present derivations for all relevant diffusion and fluorescence models, and we use simulated diffusion trajectories and photon streams to evaluate the estimator's performance. We consider simplified estimation schemes that bin the photon counts within time intervals of fixed duration, and show that they can perform well in realistic parameter regimes. The latter results indicate the feasibility of performing identification experiments in real time. It will be straightforward to generalize our approach for use in more complicated scenarios, e.g., with multiple spectral channels or fast photophysical dynamics.

Algorithms↗