Abstract
Purpose :
Optical Coherence Tomography (OCT) is a well-established modality providing cross-sectional images of the human retina noninvasively. OCT, however, does not provide high-resolution en face images of the outer retinal layers due to eye aberrations and the fundamental tradeoff between imaging depth and transverse resolution. This study develops a new imaging approach that solves those problems by providing high-resolution, aberration-free volumetric images of the retina with the voxel rate of 8 GHz.
Methods :
We developed a custom Fourier-domain full-field optical coherence tomography (FD-FF-OCT) imaging instrument based on the rapidly tunable laser, ultra-fast area scan camera, and the spatial phase modulator (SPM). The SPM is used to dynamically modulate the incident light to suppress the coherent noises from the laser. We imaged the retina of the healthy volunteer. We captured ten volumes (512x512x512 pixels), which took less than 90 ms (the time needed for the conventional scanning OCT system to capture single b-scan). Then we applied our digital adaptive optics algorithms to correct for eye aberrations in post-processing.
Results :
We imaged the retina of the healthy 44-years-old volunteer. We first applied the sensorless adaptive optics to correct the specular reflection spot from the inner retina. The initially aberrated spot was turned into the sharp 2D gaussian shaped function, whose peak intensity is approximately five times larger than the initial spot. Then, we employed the sensorless adaptive optics to the IS/OS layer. We revealed photoreceptor mosaic, which is not visible without sensorless adaptive optics nor averaging several volumes. We confirmed that observed cellular structure comprises photoreceptor cones using the power spectrum of the en face images.The power spectrum of the corrected image contains Yellot’s ring.
Conclusions :
By extending FD-FF-OCT with a spatiotemporal phase modulator (SPM), we reduce coherent noises and maintain the phase of the useful signal allowing sensorless correction of phase errors in post-processing. We employed this to depict the IS/OS layer of the human retina in vivo, revealing the photoreceptor mosaic. Our results can thus pave the way for FD-FF-OCT to in vivo cellular-level non-invasive volumetric imaging of the human retina.
This is a 2020 ARVO Annual Meeting abstract.