Automated motion-artifact correction in an OCTA image using tensor voting approach

Optical coherence tomography angiography (OCTA) is a promising tool for imaging subsurface microvascular networks owing to its micron-level resolution and high sensitivity. However, it is not uncommon that OCTA imaging suffers from strip artifacts induced by tissue motion. Although various algorithms for motion correction have been reported, a method that enables motion correction on a single en face OCTA image remains a challenge. In this study, we propose a motion correction approach based on microvasculature detection and broken gap filling. Unlike previous methods using registration to restore disturbed vasculature during motion artifact removal, tensor voting is performed in an individual projected image to connect the broken vasculature. Both simulation and in vivo 3D OCTA imaging of the mouse bladder are performed to validate the effectiveness of this method. A comparison of in vivo images before and after motion correction shows that our method effectively corrects tissue motion artifacts while preserving the continuity of vasculature network. Furthermore, in vivo results of this technique are presented to demonstrate its utility for imaging tumor angiogenesis in the mouse bladder.