Abstract
Purpose :
Hardware-based, sensorless, and computational adaptive optics optical coherence tomography (AO-OCT) have been developed to achieve cellular-resolution retinal imaging. However, limited 3D imaging fields, high cost, and intrinsic hardware complexity limit the practical utility of the existing approaches. Therefore, our goal was to demonstrate depth-invariant cellular-resolution retinal imaging over a clinically meaningful field of view (FOV) while supporting the commonly used point-scan swept-source OCT (SS-OCT) architecture.
Methods :
3-mm × 3-mm retinal FOV was scanned within 2.3 seconds of acquisition time using a phase-stable SS-OCT system based on a 1050-nm stretched-pulse mode-locked laser (SPML). The system was operating at 4.5-MHz A-scan rate with 93.6 dB sensitivity. In order to address spatially variant refractive error, the FOV was divided into 6 × 6 subvolumes each covering 600-μm × 600-μm area. Depth-independent aberration was corrected by applying a guide star iterative algorithm to the photoreceptor layer (PRL) slab. Depth-dependent defocus was removed using interferometric synthetic aperture microscopy (ISAM).
Results :
Depth-invariant cellular-resolution retinal imaging was demonstrated across 3-mm × 3-mm FOV and 9-mm × 3-mm FOV (five volumes). Cone photoreceptor cells, retinal nerve fiber layer (RNFL), and retinal capillaries were sharply visualized. Individual cone cells were resolved at as close as 0.5 mm from the foveal pit, where the cone cell density of 24,000 cones/mm was measured. Clearly resolved retinal nerve fiber layer revealed the orientations of ganglion cell axon bundles.
Conclusions :
By providing wide-field 3D cellular-resolution imaging in the human retina using a standard point-scan architecture routinely used in the clinic, this platform proposes a strategy for expanded utilization of high-resolution retinal imaging in both research and clinical settings.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.