Nucleus segmentation across imaging experiments: the 2018 Data Science Bowl

TLDR

Segmenting nuclei in microscopy images is a common first step in quantitative bioimage analysis, yet existing tools require experiment‑specific selection and configuration. The 2018 Data Science Bowl challenged participants to develop a universal segmentation method for any 2‑D light‑microscopy image of stained nuclei, without human intervention. Teams built deep‑learning models that could be applied across diverse image types and experimental conditions. Top participants succeeded, producing configuration‑free models that accurately identified nuclei across many image types and experimental conditions, marking a significant advance toward automated bioimage analysis.

Abstract

Abstract Segmenting the nuclei of cells in microscopy images is often the first step in the quantitative analysis of imaging data for biological and biomedical applications. Many bioimage analysis tools can segment nuclei in images but need to be selected and configured for every experiment. The 2018 Data Science Bowl attracted 3,891 teams worldwide to make the first attempt to build a segmentation method that could be applied to any two-dimensional light microscopy image of stained nuclei across experiments, with no human interaction. Top participants in the challenge succeeded in this task, developing deep-learning-based models that identified cell nuclei across many image types and experimental conditions without the need to manually adjust segmentation parameters. This represents an important step toward configuration-free bioimage analysis software tools.