Although cone cells are more important for everyday vision, rod cells make up the vast majority (approximately 95%) of the photoreceptors of the human eye. Rods are much more sensitive than cones. So sensitive, in fact, that they are able to respond to individual photons, giving us the ability to see in low-light conditions. And they are responsible for detecting movement in our peripheral vision.

But, owing to their small size (around 2 μm in diameter) and the optical distortions introduced by other components of the eye, resolving individual rod cells in living eyes using conventional medical imaging techniques is practically impossible. This makes it difficult to diagnose the early stages of disease in these cells in order to treat them and prevent irreversible damage.

Credit: UNIVERSITY OF ROCHESTER / BIOMEDICAL OPTICS EXPRESS

However, Alfredo Dubra and colleagues have now developed a microscope that is able to collect detailed images of the mosaic of photoreceptors in the retinas of living subjects (Biomed. Opt. Express 2, 1864–1876; 2011 and Biomed. Opt. Express 2, 1757–1768; 2011). It relies on a technique known as adaptive optics, pioneered by astronomers to correct for distortions introduced by the atmosphere and produce sharp images of the heavens using Earth-bound telescopes.

Their microscope — known as a confocal adaptive-optics scanning ophthalmascope — works in three stages: scanning a focused beam of light across the subject's retina; measuring variations in the wavefront of the reflected light, which are introduced by imperfections in the lens and cornea at the front of the eye, and then correcting for these perturbations with deformable mirrors. The result is retinal images with resolutions approaching the diffraction limit for the wavelengths of light used. Both the small cones at the centre of the retina (pictured left) and the small rods surrounding larger cones at the retina's periphery (pictured right) are clearly resolved.

Dubra et al. expect that this ability to routinely collect detailed images of retinal structures in a clinical setting will make it possible to diagnose retinal disease earlier, and to generate a wealth of previously inaccessible data for the development of better treatments.