Universal consensus 3D segmentation of cells from 2D segmented stacks
Cell segmentation is the foundation of a wide range of microscopy-based biological studies. Deep learning has revolutionized two-dimensional (2D) cell segmentation, enabling generalized solutions across cell types and imaging modalities. This has been driven by the ease of scaling up image acquisition, annotation and computation. However, three-dimensional (3D) cell segmentation, requiring dense annotation of 2D slices, still poses substantial challenges. Manual labeling of 3D cells to train broadly applicable segmentation models is prohibitive. Even in high-contrast images annotation is ambiguous and time-consuming. Here we develop a theory and toolbox, u-Segment3D, for 2D-to-3D segmentation, compatible with any 2D method generating pixel-based instance cell masks. u-Segment3D translates and enhances 2D instance segmentations to a 3D consensus instance segmentation without training data, as demonstrated on 11 real-life datasets, comprising >70,000 cells, spanning single cells, cell aggregates and tissue. Moreover, u-Segment3D is competitive with native 3D segmentation, even exceeding when cells are crowded and have complex morphologies.
Authors
- Gabriel M. Gihana (ORCID: https://orcid.org/0000-0002-5443-4440)
- Benjamin A. Nanes (ORCID: https://orcid.org/0000-0001-9686-1611)
- Reto Fiolka (ORCID: https://orcid.org/0000-0002-4636-5000)
- Md Torikul Islam (ORCID: https://orcid.org/0000-0001-8263-0628)
- Zach Marin (ORCID: https://orcid.org/0000-0001-5341-9911)
- Bo-Jui Chang (ORCID: https://orcid.org/0000-0002-5513-7106)
- Jinlong Lin (ORCID: https://orcid.org/0000-0002-7503-223X)
- Edward Jenkins (ORCID: https://orcid.org/0000-0003-3789-6421)
- Felix Zhou (ORCID: https://orcid.org/0000-0003-4463-1165)
- Gaudenz M. Danuser (ORCID: https://orcid.org/0000-0001-8583-2014)
- Clarence Yapp (ORCID: https://orcid.org/0000-0003-1144-5710)
- Stephan Daetwyler (ORCID: https://orcid.org/0000-0002-7444-4734)
- Peter Karl Sorger (ORCID: https://orcid.org/0000-0002-3364-1838)
- Andrew Raymond Jamieson (ORCID: https://orcid.org/0000-0001-5416-9379)
- Andrew Weems (ORCID: https://orcid.org/0000-0003-4193-3061)
- Sean J. Morrison (ORCID: https://orcid.org/0000-0003-1587-8329)
- Qiongjing Zou (ORCID: https://orcid.org/0000-0002-4636-0795)
- Kevin M. Dean (ORCID: https://orcid.org/0000-0003-0839-2320)
- Hazel M. Borges (ORCID: https://orcid.org/0009-0006-5834-5310)
Institutions
- Roche (Switzerland) (CH)
- University of Vienna (AT)
- Howard Hughes Medical Institute (US)
- Harvard University (US)
- Max Perutz Labs (AT)
- University of Oxford (GB)
- Southwestern Medical Center (US)
- Center for Systems Biology (US)
- The University of Texas at Austin (US)
- The University of Texas Southwestern Medical Center (US)
Publication Details
- Journal
- Nature Methods
- Published
- 2025-11-01
- DOI
- https://doi.org/10.1038/s41592-025-02887-w
- Citations
- 6
- Primary Topic
- Cell Image Analysis Techniques
- Type
- article
- Field-Weighted Citation Impact
- 9.80