To perform morphodynamic profiling from time lapse images of neurite outgrowth, we developed an edge evolution tracking (EET) algorithm, by which cell boundary movements including an arbitrary complex boundary transition are quantified. This algorithm enables us to estimate temporal evolution of cellular edge, and thus to trace the transition of any objective edge movements. We show advantages of EET by comparing it with the other two methods on an artificial data set that imitates neural outgrowth. We also demonstrate the usefulness of our EET by applying it to a data set of time-lapse imaging of neural outgrowth. The results show verification of quantitative profiling for arbitrary complex cell boundary movements.