Fast tracking of fluorescent cells based on the Chan-Vese model

We present a fast and robust approach to tracking whole fluorescent cells in time-lapse series. The proposed tracking scheme involves two steps. First, coherence-enhancing diffusion filtering is applied on each frame to reduce the amount of noise and enhance flow-like structures. Second, the enhanced cell boundaries are detected by minimizing the Chan-Vese model in a fast level set-like framework. To allow simultaneous tracking of multiple cells over time, the contour evolution has been integrated with a topological prior exploiting the object indication function. Preliminary tracking experiments on 2D time-lapse series of GFP-transfected adipose-derived stem cells demonstrate high accuracy and short execution times.