Compensating for photon counting losses in a TCSPC SPAD array enables quantitative time-resolved fluorescence anisotropy imaging.

We devise and experimentally validate a theoretical model to account for lost photon counts during the exposure time of a time-correlated single photon counting (TCSPC)-based QuantICAM single photon avalanche diode array camera. The work is motivated by the quest for TCSPC-based wide-field time-resolved fluorescence anisotropy imaging (TR-FAIM), implemented by the acquisition of images at orthogonal polarization. For accurate, quantitatively correct TR-FAIM, the two images must be acquired under equivalent conditions and any photons lost during the camera exposures must be precisely quantified. Our model is based on a binomial distribution with a single adjustable parameter. We plot the recorded versus the true photon counts for exposure times of 250 µs and 1000 µs, using photons with random arrival times and from fluorescence decays. Our model describes the experimental data well and the correct number of excitation cycles during the exposure time is extracted from least-squares fits of the binomial model to the experimental data. On the basis of this model, we account for lost photons in TCSPC-based TR-FAIM and show that a compensation for lost photons is essential to obtain quantitatively correct steady-state anisotropy andG-factor histograms in TR-FAIM. We also show that, under the conditions used, the rotational correlation time, initial anisotropyr0and hindered rotation parameterr∞histograms are only marginally affected by lost photons. Our work thus paves the way for robust and reliable TCSPC-based TR-FAIM.