Abstract
Purpose: :
High resolution adaptive optics imaging of the human retina is a potentially powerful technique for providing precise measurements of the human retina. One difficulty with adaptive optics imaging is that in general the small field of view makes it difficult to know which retinal region is being imaged, and thus it is hard to relate the detailed view to overall morphology.
Methods: :
We constructed a hybrid system that combines a wide field of view, slit–scanning confocal system, with a high resolution, fully confocal adaptive optics scanning laser ophthalmoscope. The slit scan system provides a 35 degree field of view, with a refresh rate of 15 Hz. The confocal nature of the slit scan system allows high contrast imaging, even in subjects with poor optical quality. This system also controls a set of deflection mirrors, which can steer the high resolution view (between 1 and 3 degrees) across the central 10 degrees of the wide field image using screen controls. Regions outside this 10 degree window can be imaged by changing the fixation target under computer control. The adaptive optics system uses a Boston Micromachines MEMS deformable mirror and a Shack Hartman sensor. The system uses a total of four wavelengths, one for the SH sensor (678 nm), one for high resolution imaging (830 nm), one for wide field imaging (905 nm) and the final one for retinal tracking (1060 nm).
Results: :
The system provides simultaneous high contrast retinal images at both low and high resolutions simultaneously. The actual raster location for the high resolution view is present on the wide field view, allowing direct confirmation of the region being imaged. The adaptive optics view provides robust images of cones and other structures in the retina, and can image cones, and small capillaries over most of the posterior pole.
Conclusions: :
We removed a major limitation for high resolution AO imaging, especially in subjects who may lack stable fixation: the ability to document where in the retina the images were actually obtained. While postprocessing to form larger mosaics is possible, the current system allows the experimenter to not only control the region being imaged, but to record the location of the images directly.
Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • retina