Quantitative analysis of cellular shape and motility with improved active contours

Quantitative studies of cellular motility and morphology require a method to segment and track individual cells in videomicroscopy data. We present an active contour method adapted to address important difficulties of cellular imaging. Detection of low contrast pseudopods is facilitated by using an edge map based on the average intensity deviation field. Separation of cells in contact is facilitated by introducing a repulsion effect between contours. Our method was successfully applied to assess the influence of a particular protein on the motility of the parasite Entamoeba histolytica, and in an ongoing study of chemotaxis.