Deep Learning in Image Cytometry: A Review

Gupta, Anindya; Harrison, Philip J.; Wieslander, Håkan; Pielawski, Nicolas; Kartasalo, Kimmo; Partel, Gabriele; Solorzano, Leslie; Suveer, Amit; Klemm, Anna H.; Spjuth, Ola; Sintorn, Ida‐Maria; Wählby, Carolina

doi:10.1002/cyto.a.23701

Cited by 156 publications

(118 citation statements)

References 123 publications

(150 reference statements)

Supporting

Mentioning

118

Contrasting

Order By: Relevance

“…However, we must note that training and optimizing neural networks for a certain classification task is not straight-forward. It requires significant expert knowledge to overcome obstacles such as overfitting, hyper-parameter tuning, handling big data, and dealing with a shortage of, or imbalance in labelled data (18,42). Different methods to deal with these issues, such as transfer learning or data augmentation, need to be made accessible and easy to use.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of human white blood cells using machine learning for stain-free imaging flow cytometry

Lippeveld

Knill

Ladlow

et al. 2019

Preprint

View full text Add to dashboard Cite

Imaging flow cytometry (IFC) produces up to 12 different information-rich images of single cells at a throughput of 5000 cells per second. Yet often, cell populations are still studied using manual gating, a technique that has several drawbacks. Firstly, it is hard to reproduce. Secondly, it is subjective and biased. And thirdly, it is time-consuming for large experiments. Therefore, it would be advantageous to replace manual gating with an automated process, which could be based on stain-free measurements originating from the brightfield and darkfield image channels.To realise this potential, advanced data analysis methods are required, in particular, machine learning. Previous works have successfully tested this approach on cell cycle phase classification with both a classical machine learning approach based on manually engineered features, and a deep learning approach. In this work, we compare both approaches extensively on the complex problem of white blood cell classification. Four human whole blood samples were assayed on an ImageStream-X MK II imaging flow cytometer. Two samples were stained for the identification of 8 white blood cell types, while two other sample sets were stained for the identification of resting and active eosinophils. For both datasets, four machine learning classifiers were evaluated on stain-free imagery using stratified 5-fold cross-validation. On the white blood cell dataset the best obtained results were 0.776 and 0.697 balanced accuracy for classical machine learning and deep learning, respectively. On the eosinophil dataset this was 0.866 and 0.867 balanced accuracy. From the experiments we conclude that classifying distinct cell types based on only stain-free images is possible with these techniques. However, both approaches did not always succeed in making reliable cell subtype classifications. Also, depending on the cell type, we find that even though the deep learning approach requires less expert input, it performs on par with a classical approach.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Another example is the work by Eulenberg et al (18), who developed a deep learning (DL) model, termed…”

Section: Introductionmentioning

confidence: 99%

Classification of human white blood cells using machine learning for stain-free imaging flow cytometry

Lippeveld

Knill

Ladlow

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…The U-Net itself is 50 commonly used in machine learning approaches because it is a lightweight convolutional neural 51 network (CNN) which readily captures information at multiple spatial scales within an image, 52 thereby preserving reconstruction accuracy while reducing the required number of training 53 samples and training time. U-Nets, and related deep learning approaches, have found broad 54 application to live-cell imaging tasks such as cell phenotype classification, feature 55 segmentation 10, [14][15][16][17][18][19] , and histological stain analysis [20][21][22][23] . 56 57…”

Section: Introductionmentioning

confidence: 99%

Practical Fluorescence Reconstruction Microscopy for Large Samples and Low-Magnification Imaging

LaChance

Cohen

2020

Preprint

View full text Add to dashboard Cite

Fluorescence reconstruction microscopy (FRM) is an approach where transmitted light images are passed into a convolutional neural network which then outputs predicted epifluorescence images. This approach enables many benefits including reduced phototoxicity, freeing up of fluorescence channels, simplified sample preparation, and the ability to re-process legacy data for new insights. However, current FRM benchmarks are single scores that are difficult to relate to how useful for trustworthy and FRM predicition is. Here, we relate the conventional benchmark to practical and familiar cell biology analyses to demonstrate that FRM should be judged in context. We further demonstrate that it performs remarkably well even with lowermagnification microscopy data, as are often collected in high content imaging. Specifically, we present promising results for nuclei, cell-cell junctions, and fine feature reconstruction; provide experimental design guidelines; and provide the code, sample data, and user manual to enable more widespread adoption of FRM. where S = Scaling B/ Scaling A Ex:S Zeiss->Nikon = 1.4

show abstract

“…Additionally, feature extraction from microscopy images is a highly variable process that depends on many manually tuned parameters. In order to address these issues, machine learning approaches have been used to phenotype cells directly using microscopy images 8,9 . However, previous phenotyping studies have been restricted to broad cell groups, such as lymphocytes, granulocytes, and erythrocytes, that have morphologies easily distinguishable to the human eye [10][11][12][13] .…”

Section: Introductionmentioning

confidence: 99%

Image-based Cell Phenotyping Using Deep Learning

Berryman

Matthews

Lee

et al. 2019

Preprint

View full text Add to dashboard Cite

The ability to phenotype cells is fundamentally important in biological research and medicine. Current methods rely primarily on fluorescence labeling of specific markers. However, there are many situations where this approach is unavailable or undesirable. Machine learning has been used for image cytometry but has been limited by cell agglomeration and it is unclear if this approach can reliably phenotype cells indistinguishable to the human eye. Here, we show disaggregated single cells can be phenotyped with a high degree of accuracy using low-resolution bright-field and non-specific fluorescence images of the nucleus, cytoplasm, and cytoskeleton. Specifically, we trained a convolutional neural network using automatically segmented images of cells from eight standard cancer cell-lines. These cells could be identified with an average classification accuracy of 94.6%, tested using separately acquired images. Our results demonstrate the potential to develop an "electronic eye" to phenotype cells directly from microscopy images indistinguishable to the human eye.Image Based Cell Phenotyping Using Deep-Learning

show abstract

Deep Learning in Image Cytometry: A Review

Cited by 156 publications

References 123 publications

Classification of human white blood cells using machine learning for stain-free imaging flow cytometry

Classification of human white blood cells using machine learning for stain-free imaging flow cytometry

Practical Fluorescence Reconstruction Microscopy for Large Samples and Low-Magnification Imaging

Image-based Cell Phenotyping Using Deep Learning

Contact Info

Product

Resources

About