Image formation theory of optical coherence tomography with optical aberrations and its application to computational aberration correction.

Computational corrections of defocus and aberrations in optical coherence tomography (OCT) offers a promising approach to realize high-resolution imaging with deep imaging depth, but without additional high hardware costs. However, these techniques are not well understood owing to a lack of accurate theoretical models and investigation tools. The image formation theory for OCT with optical aberrations is thus formulated here. Based on this theory, a numerical simulation method is developed, and computational refocusing and computational aberration correction (CAC) methods are designed. The CAC method based on the image formation theory is applied to simulated OCT signals and OCT images of a microparticle phantom and an in vivo human retina for simultaneous multi-depth correction of systematic aberration. The numerical simulation under the effective numerical aperture of 0.2 and 1.05 µm central wavelength shows that the proposed method can obtain the Strehl ratios of more than 0.8 over a ± 100 µm defocus range, while the conventional method cannot achieve this under the simulated conditions. Imaging results show that the CAC method designed based on the image formation theory can correct optical aberrations and improve the image quality more than the conventional CAC method. The proposed method improved the frequency component corresponding to the density of cone photoreceptors in OCT photoreceptor images by 1.2 to 1.4 times under the multi-depth correction. This theoretical model-based approach provides a powerful aid for understanding OCT imaging properties and processing method design.