The goal of the 2018 DSB was to segment the nucleus of cells in a variety of microscopy images. In contrast to previous cell segmentation challenges that requiere methods to be tuned for image sets separately, this competition was focused on having a single segmentation model that can blindly segment nuclei in 2D light microscopy images, without user interaction. This repository contains code used for analyzing the result of the competition. A manuscript describing the findings will be posted here soon.
The top three participants, according to the official leaderboard, have made their solutions publicly available in the following repositories: