VollSeg Extensions and Workflow Development with User Support

VollSeg is a deep learning-based 2D and 3D segmentation tool for irregularly shaped cells. It was originally developed for the segmentation of densely packed membrane labeled cells in challenging images with low signal-to-noise ratios. 

The current VollSeg plugin for napari lacks a module to create training data that incorporates choice of augmentation, morphological operations required as pre-processing step in training such models, and segmentation metrics to evaluate the model performance. Post segmentation, the output lacks the workflow that would provide end users with direct reads for their experiments, such as tracking cells or tracking tissue branch morphogenesis using skeletonization, performing RANSAC based fits on kymographs to extract dynamic instability parameters for microtubules.

This project will enhance vollseg-napari by providing documentation, trained models, and a model training module to integrate it with existing napari plugins for tracking, skeletonization, RANSAC fits and segmentation metrics.