A novel method for multiple lifetime recovery in time-domain optical imaging

摘要

A new multiple lifetime fitting algorithm is presented which deconvolves a time-domain system Instrument Response Function (IRF) from a measured Fluorescence Time Point Spread Function (FTPSF) prior to lifetime fitting. Deconvolution is followed by filtering, using a special case of the optimal Wiener filter, where the signal-to-noise ratio (SNR) in the spectral domain is evaluated empirically, and thus tuned with respect to each specific FTPSF-IRF combination at hand. Comparisons between the proposed deconvolution scheme and the classical Iterative Convolution (IC) scheme over a set of simulated and experimental data reveal that the proposed scheme typically exhibits order-of-magnitude performance gains (accuracy and efficiency combined) over the IC scheme in realistic conditions.