Automated Deep Learning-Based System to Identify Endothelial Cells Derived from Induced Pluripotent Stem Cells

Kusumoto, Dai; Lachmann, Mark; Kunihiro, Takeshi; Yuasa, Shinsuke; Kishino, Yoshikazu; Kimura, Mai; Katsuki, Toshiomi; Itoh, Shogo; Seki, Tomohisa; Fukuda, Keiichi

doi:10.1016/j.stemcr.2018.04.007

Cited by 87 publications

(76 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has become increasingly popular in finding latent data structures and classifying highly nonlinear datasets, attributes that are ideal for physical sciences and, more particularly, photonics 49 . Machine learning has also proven ideal for classifying cell subtypes from label-free images 85 to predict macrophage activation, 86 lymphocyte cell types, 87 pluripotent stem cell-derived endothelial cells 88 and differentiating primary hematopoietic progenitors 89 . Our analysis using a pre-trained MLP neural network clearly demonstrated that supervised machine learning can be successfully deployed to classify cells from normal and injured vessels with autofluorescence intensity as the only input.…”

Section: Discussionmentioning

confidence: 99%

“…Multi-layered neural networks that mimic a human neural circuit structure have become increasingly popular in finding latent data structures and classifying highly nonlinear photonic datasets [29]. They have also proven ideal for classifying cell subtypes from label-free images [49] to predict macrophage activation, [50] lymphocyte cell types, [51] pluripotent stem cellderived endothelial cells [52], and differentiating primary hematopoietic progenitors [53].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Disease-relevant single cell photonic signatures identify S100β stem cells and their myogenic progeny in vascular lesions

Molony

King

Luca

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: A hallmark of subclinical atherosclerosis is the accumulation of vascular smooth muscle cell (SMC)-like cells leading to intimal thickening, lipid retention and plaque formation, yet their origin remains controversial. The ability to discriminate heterogeneous populations of cells in the context of various disease processes is of potential clinical and diagnostic value. Methods:The feasibility of multivariate analysis of single cell photonics as a discriminator of cell phenotype was assessed using label-free optical multi-parameter interrogation of single cells on a novel Lab-on-a-Disk (LoaD) platform before myogenic differentiation of a series of multipotent stem cells in vitro, and the cellular heterogeneity within vascular lesions in vivo, was determined using supervised machine learning and validated by genetic lineage tracing in vivo.Findings: Single cell photonics were of sufficient coverage, specificity, and quality to discriminate various disparate cell phenotypes in vitro and normal medial SMCs from SMClike cells following injury ex vivo, in addition to distinguishing myogenic differentiation of a series of multipotent stem cells in vitro. Supervised machine learning of these photonic datasets supported genetic lineage tracing analysis and identified the presence of S100β + stem-derived myogenic progeny within vascular lesions, in part, due to upregulation of Coll 3A1 and elastin.Interpretation: Disease-relevant photonic signatures reflect cell type and differentiation state and may have important predictive value as a phenotypic discriminator for vascular disease. Research in contextEvidence before this study: Cardiovascular disease (CVD), the leading cause of death and disability world-wide is characterized by pathological structural changes to the blood vessel wall. Pathologic observations in humans vessels confirm that early 'transitional' lesions enriched with SMC-like cells are routinely present in atherosclerotic-prone regions of arteries during pathologic intimal thickening, prior to lipid retention, and the appearance of a atherosclerotic plaque. Lineage tracing and single cell RNA sequence analysis (scRNA-seq) analysis has provided compelling evidence for the involvement of differentiated medial SMC and adventitial/medial progenitors derived from SMCs in progressing intimal thickening.Despite these insights, the putative role of resident vascular stem cells that do not originate from medial SMCs in promoting intimal thickening remains controversial. Light as a diagnostic and prognostic tool has several advantages including high sensitivity, non-destructive measurement, small or even non-invasive analysis and low limits of detection for early detection of disease phenotypes. In combination with microfluidics, photonics enables real time measurement of single cells in very small sample volumes. To accompany these photonic platforms, deep learning, a subset of supervised machine learning based primarily on artificial neural network geometries, has rapidly grown as a predictive method t...

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Disease-relevant single cell photonic signatures identify S100β stem cells and their myogenic progeny in vascular lesions

Molony

King

Luca

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…CNNs have also shown the capacity to accurately identify cellular states from simple transmitted-light data. For example, differentiating cells based on cell cycle stage (36), cells affected by phototoxicity (37) or stem cell-derived endothelial cells (38). Previously, researchers would generally rely in fluorescent reporters to identify these cellular features, CNNs now enable the same findings in a label-free manner.…”

Section: Cnns In Microscopymentioning

confidence: 99%

Artificial Intelligence for Microscopy: What You Should Know

Chamier

Laine

Henriques

2019

Preprint

View full text Add to dashboard Cite

Artificial Intelligence based on Deep Learning is opening new horizons in Biomedical research and promises to revolutionize the Microscopy field. Slowly, it now transitions from the hands of experts in Computer Sciences to researchers in Cell Biology. Here, we introduce recent developments in Deep Learning applied to Microscopy, in a manner accessible to non-experts. We overview its concepts, capabilities and limitations, presenting applications in image segmentation, classification and restoration. We discuss how Deep Learning shows an outstanding potential to push the limits of Microscopy, enhancing resolution, signal and information content in acquired data. Its pitfalls are carefully discussed, as well as the future directions expected in this field. Artificial intelligence | Machine learning | Live-cell imaging | Super-resolution microscopy | Classification | SegmentationCorrespondence: r.laine@ucl.ac.uk, r.henriques@ucl.ac.uk

show abstract

“…These methods are automatic, have high reproducibility, and are non-invasive to cells because fluorescent markers are not needed [15]. Convolutional neural networks (CNNs), which are deep learning-based networks, are capable can achieve extraordinary outcomes when applied to image analysis and identification tasks [16,17]. For example, for the identification of differentiated and undifferentiated endothelial cells, Kusumoto et al have developed a CNN-based algorithm that has an accuracy of 91.3% and an F-measure (weighted harmonic mean of the precision and the recall of the classification) of 80.9%.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning for Non-Invasive Determination of the Differentiation Status of Human Neuronal Cells by Using Phase-Contrast Photomicrographs

et al. 2019

View full text Add to dashboard Cite

Regenerative medicine using neural stem cells (NSCs), which self-renew and have pluripotency, has recently attracted a lot of interest. Much research has focused on the transplantation of differentiated NSCs to damaged tissues for the treatment of various neurodegenerative diseases and spinal cord injuries. However, current approaches for distinguishing differentiated from non-differentiated NSCs at the single-cell level have low reproducibility or are invasive to the cells. Here, we developed a fully automated, non-invasive convolutional neural network-based model to determine the differentiation status of human NSCs at the single-cell level from phase-contrast photomicrographs; after training, our model showed an accuracy of identification greater than 94%. To understand how our model distinguished between differentiated and non-differentiated NSCs, we evaluated the informative features it learned for the two cell types and found that it had learned several biologically relevant features related to NSC shape during differentiation. We also used our model to examine the differentiation of NSCs over time; the findings confirmed our model’s ability to distinguish between non-differentiated and differentiated NSCs. Thus, our model was able to non-invasively and quantitatively identify differentiated NSCs with high accuracy and reproducibility, and, therefore, could be an ideal means of identifying differentiated NSCs in the clinic.

show abstract

Automated Deep Learning-Based System to Identify Endothelial Cells Derived from Induced Pluripotent Stem Cells

Cited by 87 publications

References 28 publications

Disease-relevant single cell photonic signatures identify S100β stem cells and their myogenic progeny in vascular lesions

Disease-relevant single cell photonic signatures identify S100β stem cells and their myogenic progeny in vascular lesions

Artificial Intelligence for Microscopy: What You Should Know

Deep Learning for Non-Invasive Determination of the Differentiation Status of Human Neuronal Cells by Using Phase-Contrast Photomicrographs

Contact Info

Product

Resources

About