Edge detection at multiple locations using a ‘radar’ tracking algorithm as exemplified in isolated guinea-pig lymphatic vessels

An algorithm based on the Kalman filter and used for tracking a target in a noisy environment has been applied to monitoring the position of moving boundaries as recorded by a video camera. The method, termed the Probabilistic Data Association Filter, is computationally efficient, allowing simultaneous monitoring at many positions along the boundary at video frame rates of 25 Hz. The present study uses the tracking procedure to investigate and correlate constrictions along lymphatic vessels. The procedure is well suited for real time tracking of visualised biological events including the simultaneous monitoring of constrictions in networks of blood vessels, the movement of cells or the trajectory of intracellular events monitored by intracellular fluorescent markers.