Cellpose 2.0: how to train your own model

Pachitariu, Marius; Stringer, Carsen

doi:10.1038/s41592-022-01663-4

Cited by 441 publications

(446 citation statements)

References 50 publications

Supporting

Mentioning

326

Contrasting

Order By: Relevance

“…The choice of FogBank was based on findings by Vicar et al (2019) that compared it favorably against a number of common contrast microscopy segmentation methods. CellPose is a supervised algorithm for cell segmentation trained on a wide array of community-supplied fluorescent and non-fluorescent microscopy images ( Pachitariu and Stringer, 2022 ; Stringer et al , 2021 ). For comparison, we define two metrics of performance: binary error,

, where A gt and A pred are the foreground area of the ground truth segmentation and predicted segmentation, respectively, and object count error,

, where N gt and N pred are the number of cells in the ground truth segmentation and predicted segmentation, respectively.…”

Section: Resultsmentioning

confidence: 99%

Biophysical informatics reveals distinctive phenotypic signatures and functional diversity of single-cell lineages

2022

View full text Add to dashboard Cite

Motivation In this work, we present an analytical method for quantifying both single-cell morphologies and cell network topologies of tumor cell populations and use it to predict 3D cell behavior. Results We utilized a supervised deep learning approach to perform instance segmentation on label-free live cell images across a wide range of cell densities. We measured cell shape properties and characterized network topologies for 136 single-cell clones (SCCs) derived from the YUMM1.7 and YUMMER1.7 mouse melanoma cell lines. Using an unsupervised clustering algorithm, we identified six distinct morphological subclasses. We further observed differences in tumor growth and invasion dynamics across subclasses in an in vitro 3D spheroid model. Compared to existing methods for quantifying 2D or 3D phenotype, our analytical method requires less time, needs no specialized equipment, and is capable of much higher throughput, making it ideal for applications such as high-throughput drug screening and clinical diagnosis. Availability and Implementation //github.com/trevor-chan/Melanoma_NetworkMorphology. Supplemental Information Bioinformatics online.

show abstract

, where A gt and A pred are the foreground area of the ground truth segmentation and predicted segmentation, respectively, and object count error,

, where N gt and N pred are the number of cells in the ground truth segmentation and predicted segmentation, respectively.…”

Section: Resultsmentioning

confidence: 99%

Biophysical informatics reveals distinctive phenotypic signatures and functional diversity of single-cell lineages

2022

View full text Add to dashboard Cite

show abstract

“…Accordingly, we consider AimSeg as a powerful tool to train novel deep learning models. Indeed, amending segmentation results obtained automatically has proven to be an efficient approach to improve pre-existing models [47].…”

Section: Discussionmentioning

confidence: 99%

AimSeg: a machine-learning-aided tool for axon, inner tongue and myelin segmentation

Rondelli

Morante‐Redolat

Bankhead

et al. 2023

Preprint

View full text Add to dashboard Cite

Electron microscopy (EM) images of axons and their ensheathing myelin from both the central and peripheral nervous system are used for assessing myelin formation, degeneration (demyelination) and regeneration (remyelination). The g-ratio is the gold standard measure of assessing myelin thickness and quality, and traditionally is determined from measurements done manually from EM images – a time-consuming endeavour with limited reproducibility. These measurements have also historically neglected the innermost uncompacted myelin sheath, known as the inner myelin tongue. Nonetheless, the inner tongue has been shown to be important for myelin growth and some studies have reported that certain conditions can elicit its enlargement. Ignoring this fact may bias the standard g-ratio analysis, whereas quantifying the uncompacted myelin has the potential to provide novel insights in the myelin field. In this regard, we have developed AimSeg, a bioimage analysis tool for axon, inner tongue and myelin segmentation. Aided by machine learning classifiers trained on tissue undergoing remyelination, AimSeg can be used either as an automated workflow or as a user-assisted segmentation tool. Validation results show good performance segmenting all three fibre components, with the assisted segmentation showing the potential for further improvement with minimal user intervention. This results in a considerable reduction in time for analysis compared with manual annotation. AimSeg could also be used to build larger, high quality ground truth datasets to train novel deep learning models. Implemented in Fiji, AimSeg can use machine learning classifiers trained in ilastik. This, combined with a user-friendly interface and the ability to quantify uncompacted myelin, makes AimSeg a unique tool to assess myelin growth.

show abstract

“…All the cells were inspected. Alternatively, for rabbit and monkey embryos, we used a custom-made cell detection and tracking algorithm (unpublished) based on nuclear segmentation with Cellpose (Pachitariu and Stringer, 2022; Stringer et al, 2021) and semi-automatic shape tracking using Napari (Sofroniew, Nicholas et al, 2022).…”

Section: Methodsmentioning

confidence: 99%

Temporal variability and cell mechanics control robustness in mammalian embryogenesis

Fabrèges

Corominas-Murtra

Moghe

et al. 2023

Preprint

View full text Add to dashboard Cite

How living systems achieve precision in form and function despite their intrinsic stochasticity is a fundamental yet open question in biology. Here, we establish a quantitative morphomap of pre-implantation embryogenesis in mouse, rabbit and monkey embryos, which reveals that although blastomere divisions desynchronise passively without compensation, 8-cell embryos still display robust 3D structure. Using topological analysis and genetic perturbations in mouse, we show that embryos progressively change their cellular connectivity to a preferred topology, which can be predicted by a simple physical model where noise and actomyosin-driven compaction facilitate topological transitions lowering surface energy. This favours the most compact embryo packing at the 8- and 16-cell stage, thus promoting higher number of inner cells. Impairing mitotic desynchronisation reduces embryo packing compactness and generates significantly more cell mis-allocation and a lower proportion of inner-cell-mass-fated cells, suggesting that stochasticity in division timing contributes to achieving robust patterning and morphogenesis.

show abstract

Cellpose 2.0: how to train your own model

Cited by 441 publications

References 50 publications

Biophysical informatics reveals distinctive phenotypic signatures and functional diversity of single-cell lineages

Biophysical informatics reveals distinctive phenotypic signatures and functional diversity of single-cell lineages

AimSeg: a machine-learning-aided tool for axon, inner tongue and myelin segmentation

Temporal variability and cell mechanics control robustness in mammalian embryogenesis

Contact Info

Product

Resources

About