Abstract
Estimation of signal parameters via rotational invariance techniques is a classical algorithm widely used in array signal processing for direction-of-arrival estimation of emitters. Inspired by this method, a new signal model and new fluorescence lifetime estimation via rotational invariance techniques (FLERIT) were developed for multiexponential fluorescence lifetime imaging (FLIM) experiments. The FLERIT only requires a few time bins of a histogram generated by a time-correlated single-photon counting FLIM system, greatly reducing the data throughput from the imager to the signal processing units. As a noniterative method, the FLERIT does not require initial conditions, prior information nor model selection that are usually required by widely used traditional fitting methods, including nonlinear least square methods or maximum-likelihood methods. Moreover, its simplicity means it is suitable for implementations in embedded systems for real-time applications. FLERIT was tested on synthesized and experimental fluorescent cell data showing the potentials to be widely applied in FLIM data analysis.
Highlights
F LUORESCENCE lifetime imaging (FLIM) is a powerful tool to study the microstructure and microenvironments of molecules
Inspired by ESPRIT, we proposed a new signal model and applied this model for estimation of fluorescence lifetime based on rotational invariance techniques, a system we term fluorescence lifetime estimation via rotational invariance techniques (FLERIT)
We proposed a new method called FLERIT
Summary
F LUORESCENCE lifetime imaging (FLIM) is a powerful tool to study the microstructure and microenvironments of molecules. Despite the potential and significant impact of FLIM, primarily in the biological sciences, estimation of the fluorescence lifetimes remains a significant challenge, with low photon counts systems such as in rapid live cell imaging. This is becoming increasingly demanding with the development of novel CMOS SPAD arraybased widefield FLIM systems, which can generate significant volumes of data [9]–[11]. We propose a new method of lifetime extraction based on a classical algorithm called estimation of signal parameters via rotational invariance techniques (ESPRIT) [22]. This paper presents the theory (see Section II), and demonstrates the potential through application to both simulated (see Section III) and experimental (Section IV) data
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
