Highly efficient eigen decomposition based statistical optical microangiography

Optical coherence tomography (OCT) (1,2)/Optical coherence tomography angiography (OCTA) (3) has recently drawn increasing attention in the field of biomedical imaging due to its rapid clinical translation to ophthalmic imaging within 1 year or so. With a goal to contrast blood flow within microcirculatory tissue beds, there are a number of data processing methods available that can achieve OCTA. Most published methods use the magnitude of the OCT signal, such as speckle variance (4), correlation mapping (5), and split-spectrum amplitude decorrelation angiography (6). Some other methods use the phase information, such as phase variance OCT (7). Leveraging the complete information available in the OCT system, the complex OCT signal is explored, such as optical microangiography (OMAG) (8,9), complex differential variance (CDV) (10), and eigen-decomposition (ED) based OMAG (11). Among them, ED approach is a statistical analysis method that utilizes the statistical properties of time varying complex OCT signals to achieve the purpose of contrasting blood flow within tissue. ED analysis belongs to adaptive filtering category in statistical signal processing discipline (12,13), which has been widely used in magnetic resonance imaging (14) and ultrasonic imaging (15) where it serves to analyze dynamic signals from the measured signals in patients