Purchase this article with an account.
Peng XIAO, Viacheslav Mazlin, Kate Grieve, Jose A. Sahel, Mathias Fink, Claude Boccara; In vivo cellular resolution imaging of human retina with full-field OCT. Invest. Ophthalmol. Vis. Sci. 2018;59(9):278. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We have previously presented full-field OCT (FFOCT) images of ex vivo retinal tissues (Grieve et al, IOVS 2016; Thouvenin et al, IOVS 2017) resolving cellular retinal structures. In this study we apply FFOCT with spatially incoherent illumination, for which the spatial resolution is insensitive to eye aberrations, to in vivo high resolution en face human retinal imaging without using adaptive optics (AO).
An FFOCT system for retinal imaging was combined with a spectral-domain OCT (SDOCT) system. A healthy subject without pupil dilation was imaged at both the retinal periphery and the fovea. Real-time optical path length matching to image various retinal layers was achieved using the SDOCT cross-sectional images to provide the depth location of FFOCT en face images. FFOCT retinal images were acquired at 200Hz, fast enough to freeze eye motion during image acquisition. No hardware or computational AO was applied. Image stacks acquired within 200ms were registered and averaged to improve signal to noise ratio. FFOCT images were compared with AO fundus camera images.
The FFOCT images of the retinal nerve fiber layer (RNFL) and the photoreceptor inner/outer segment (IS/OS) junction layer at retinal near periphery (6°) clearly revealed structural information such as nerve fiber orientation, blood vessel distribution, and the photoreceptor mosaic. The photoreceptor mosaic spacing in FFOCT images agreed with the literature and with images acquired with an AO fundus camera. Images of the foveal IS/OS layer also showed strong signals from cone photoreceptors although the full foveal cone mosaic was not resolved when imaging with a non-dilated pupil.
We have successfully achieved FFOCT retinal imaging in the human eye in vivo without AO. This is the first time to the best of our knowledge that cellular retinal structures like never fiber bundles and the photoreceptor mosaic are imaged without using AO or wavefront post-processing. This opens the way for a straightforward implementation of a compact FFOCT system for high resolution en face retinal imaging in clinical studies.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
In vivo FFOCT human retinal images compared with AO fundus camera. (a,b) FFOCT images of the RNFL (a) and IS/OS layer (b) at the retinal near periphery (6°); (c) at the fovea center; (d,e) 2D power spectra of (b,c); (f-h) AO fundus camera images of the same locations as (a-c); (i,j) 2D power spectra of (g,h). Scale bar: 100 µm.
This PDF is available to Subscribers Only