Quantitative Inference of Bacterial Motility Behavior

This paper proposes a framework for automatic bacteria motile trajectories detection and motility behavior clustering. The input data is a sequence of images which contains bacteria motility information. The traditional experimental methods to identify the trajectories, and segment them to "run" and "tumble" modes are time consuming and subjective. The proposed method processes bacteria motility movies and extracts statistical features of runs and tumbles which drastically saves time, human labor, and minimizes human error. The statistics will be used in simulations to model bacteria motility. The methodology can be replicated with similar format of experimental microscopic images.