Abstract
Purpose :
The implementation of adaptive optics fluorescence lifetime imaging ophthalmoscopy (AOFLIO) in the living human eye employing a raster scanning imaging regime necessitates collecting >1000 autofluorescence photons at each pixel on a 2D frame. This process encounters challenges from nonlinear sampling induced by resonant scanning of the fast scanner and the intra- and inter-frame displacements caused by rapid eye movement. We present strategic approaches to ensure the precise measurement of retinal autofluorescence lifetime in the living human eye at the cellular level.
Methods :
An AOFLIO instrument was employed to measure the retinal autofluorescence lifetime in the living human eye using the time-correlated single photon counting method (excitation λ = 473 nm, autofluorescence detection in 500 – 560 nm and 560 – 720 nm spectral channels). It measures the time differences between emission photons and the corresponding excitation laser pulses and checks the arriving autofluorescence photons at a frequency of 40 MHz and marks the detected photon with a ‘timestamp’ corresponding to the spatiotemporal position of the raster scan. Meanwhile, it simultaneously records high-contrast adaptive optics confocal images of cone photoreceptors with AOFLIO to measure eye movement. An image dewarping algorithm was programmed to redistribute the AOFLIO pixels through a non-interpolative approach to linearize the sinusoidal distortion induced by the resonant scanner. A substrip-based image registration algorithm was used to calculate the retinal movements, enabling high-precision assignment of the autofluorescence photons to the correct spatiotemporal locations, that is, the pixel positions in consecutive frames. The AOFLIO detects autofluorescence photons four times at each pixel and renders a fluorescence lifetime image with a field of view of 1.2° × 1.2° in 512 × 512 pixels. The fluorescence decay at each pixel was modeled with a tri-exponential function to determine the fluorescence lifetimes and weights of different fluorescent components.
Results :
AOFLIO produced high-resolution retinal autofluorescence lifetime images disclosing clear retinal pigment epithelium cells in the human eye.
Conclusions :
High-precision photon registration ensured precise measurement of the retinal autofluorescence lifetime at the cellular level, facilitating in vivo study of the structure and metabolic function of the retina and RPE in the human eye.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.