Imaging and graphing of cortical vasculature using dynamically focused optical coherence microscopy angiography

Recently, optical coherence tomography (OCT) angiography has enabled label-free imaging of vasculature based on dynamic scattering in vessels. However, quantitative volumetric analysis of the vascular networks depicted in OCT angiography data has remained challenging. Multiple-scattering tails (artifacts specific to the imaging geometry) make automated assessment of vascular morphology problematic. We demonstrate that dynamically focused optical coherence microscopy (OCM) angiography with a high numerical aperture, chosen so the scattering length greatly exceeds the depth-of-field, significantly reduces the deleterious effect of multiple-scattering tails in synthesized angiograms. Capitalizing on the improved vascular image quality, we devised and tailored a self-correcting automated graphing approach that achieves a reconstruction of cortical microvasculature from OCM angiography data sets with accuracy approaching that attained by trained operators. The automated techniques described here will facilitate more widespread study of vascular network topology in health and disease.