Depth-resolved optimization of a real-time sensorless adaptive optics optical coherence tomography

Abstract

Sensorless adaptive optics optical coherence tomography (AO-OCT) is a technology to image retinal tissue with high resolution by compensating ocular aberrations without wavefront sensors. In this Letter, a fast and robust hill-climbing algorithm is developed to optimize five Zernike modes in AO-OCT with a numerical aperture between that of conventional AO and commercial OCT systems. The merit function is generated in real time using graphics processing unit while axially tracking the retinal layer of interest. A new method is proposed to estimate the largest achievable field of view for which aberrations are corrected uniformly in sensorless AO-OCT.