Adaptive optics is a technology developed for astronomic imaging but also currently used to image the retina of the eye. Traditional optics are not effective at resolving the image at the scale needed, as optical aberrations blur the image. AO solves this by combining a device that measures the aberrations with a mirror that deforms in shape to correct for optical aberrations. In doing so, it returns a clear image of the retina. It can be used for both research and ophthalmological diagnosis of clinical conditions. The system is not a complete device suite but instead is being built in collaboration with established AO labs at UC Berkeley.
The AOSLO system combines a 3-channel scanning laser ophthalmoscope system (SLO) with an additional channel for adaptive optics (AO) technology and is based on system designs currently in Berkeley in the lab of Dr. Austin Roorda, and the lab of Dr. Ramkumar Sabesan at the University of Washington. The system provides the ability to image, stimulate, and identify individual photoreceptors in the intact human eye. The system is also confocal, allowing the imaging of additional retinal elements at different retinal depths as the focal plane is adjusted. The system utilizes a super-continuum laser source to provide source light at multiple wavelengths (currently 543, 680, 830, 940 nm). The SLO portion of the system uses a system of lenses and mirrors to produce a tiny beam that is scanned across the retina. The small amount of light reflected from the eye is confocally imaged on pinholes at the entrance of three detectors (photomultiplier tubes), one for each wavelength. The changes in intensity of light at the detector over time is computed for each position of the scanning mirrors and video images are reconstructed from these data. The major components of the AO portion include a light source (the 940 nm channel from the supercontinuum laser), a way to measure optical imperfections/aberrations (Shack-Hartmann wavefront sensor), and a way to correct the image (computer-controlled deformable mirror). Image imperfections are constantly monitored and corrected in a closed-loop at 30Hz). There are also provisions in the system and software to monitor and correct for eye movements using n feature detection algorithm, providing image stabilization and eye-tracking at cellular scales. This property allows for stimulation of individual cones in a patch for extended periods of time if required. The subject’s eye position and pupil can be monitored with a CCD camera aimed at the eye. Finally, there is an option to add in a projector channel that provides the capability to project images, patterns, or backgrounds when designing experiments.