Fusion of Hungarian and Hierarchical Kalman Tracking Methods in Ultrasound Localization Microscopy

Ultrasound Localization Microscopy (ULM) can map microvasculature and microcirculation with unprecedented resolution by tracking isolated microbubbles (MB). The common MB tracking methods in ULM, such as the Hungarian algorithm or nearest-neighbor approaches, are fast and robust. However, while Kalman filter methods enable robust MB tracking and improve blood flow speed measurements, they often fail to reveal detailed and low-velocity microvasculature and microcirculations. This study hypothesizes that the fusion of MB trajectories from Hungarian and Kalman filter-based methods will enhance the visualization of the density and velocity images in ULM. ULM results for a rat kidney were compared across the tracking methods. The Hungarian tracker effectively captures dense vascular networks, while the Kalman tracker provides stable velocity estimates. Their combination leverages the strengths of both methods, revealing more continuous and detailed microvascular structures in both density and velocity maps compared to either tracker alone.