The analysis of OMAG in the slow axis has been known as ultrahigh sensitive OMAG (UHS-OMAG),24 which is sensitive to the slow flow velocities found in capillary vessels. In this method, the signal processing is done between A-lines across different B-frames, which makes it possible not to compromise between the fast imaging speed advantage in FD-OCT and the requirement of a relatively long time interval between A-lines to image smaller vessels, e.g., capillary vessels. To obtain the signals that represent the blood flow, a high-pass filter is used as a differential operation on the complex OCT signals, , to eliminate the static tissue signals. Display Formula
(1)where represents the index of the B-scan in the slow scan direction, and is the location of the A-line in the B-scan. The magnitude of differentiated signals in Eq. (1) represents the strength of the flow signals proportional to the particle concentration within the interrogated focal volume.25,26 In this implementation, typically a system with a 300 Hz frame rate is enough to achieve the necessary sensitivity to image slow flows in capillaries because it has been demonstrated that such a system has a sensitivity of , which is sufficient to image particle Brownian motion.24 Another variation of the OMAG scanning pattern is to perform repeated B-scans (typically two or more) at one transverse location (i.e., fast axis), then move to the next transverse location to do the same until the last location is reached in the slow scan axis.27 At each location, the analysis of Eq. (1) is applied to the repeated B-scans to extract the motion signal due to moving particles, for example, red blood cells (RBCs). A similar algorithm was also proposed by Srinivasan et al.,28 in which the angiogram was obtained by taking the difference of the weighted OCT signals acquired between adjacent B-scans.