Cell Tracking in 3D using deep learning segmentations

Live-cell imaging is a highly used technique to study cell migration and dynamics over time. Although many computational tools have been developed during the past years automatically detect track cells, they are optimized nuclei with similar shapes and/or cells not clustering together. These existing challenged when tracking fluorescently labelled membranes of due cell's irregular shape, variability in size dynamic movement across Z planes making it difficult them. Here we introduce detailed analysis pipeline perform segmentation accurate shape information, combined BTrackmate, customized codebase popular ImageJ/Fiji software Trackmate, inside tissue interest. We VollSeg, new method able membrane-labelled low signal-to-noise ratio dense packing. Finally, also created an interface Napari, Euler angle based viewer, visualize tracks along chosen view possible follow plane motion. Importantly, provide protocol implement this dataset, together required Jupyter notebooks. Our codes open source available at Github2021.