Segment and track neurons in 3D by repulsive snake method

We present a snake (active contour) model based on repulsive force to segment neurons obtained from microscopy. Based on these segmentation results, we track the neurons in 3D image to look for its branch structure. These segmentation results allow user to study morphology of neurons to further investigate neuronal function and connectivity. This repulsive snake model can successfully segment two or multiple neurons that are close to each other by some alternating repulsive force generated from the neighboring objects. We apply our results on real data to demonstrate the performance of our method.