Go to JCI Insight
  • About
  • Editors
  • Consulting Editors
  • For authors
  • Alerts
  • Advertising/recruitment
  • Subscribe
  • Contact
  • Current Issue
  • Past Issues
  • By specialty
    • Cardiology
    • Gastroenterology
    • Immunology
    • Metabolism
    • Nephrology
    • Neuroscience
    • Oncology
    • Pulmonology
    • Vascular biology
    • All...
  • Videos
    • Conversations with Giants in Medicine
    • Author's Takes
  • Reviews
    • View all reviews...
    • Mechanisms Underlying the Metabolic Syndrome (Oct 2019)
    • Reparative Immunology (Jul 2019)
    • Allergy (Apr 2019)
    • Biology of familial cancer predisposition syndromes (Feb 2019)
    • Mitochondrial dysfunction in disease (Aug 2018)
    • Lipid mediators of disease (Jul 2018)
    • Cellular senescence in human disease (Apr 2018)
    • View all review series...
  • Collections
    • Recently published
    • In-Press Preview
    • Commentaries
    • Concise Communication
    • Editorials
    • Viewpoint
    • Scientific Show Stoppers
    • Top read articles
  • Clinical Medicine
  • JCI This Month
    • Current issue
    • Past issues

  • About
  • Editors
  • Consulting Editors
  • For authors
  • Current issue
  • Past issues
  • By specialty
  • Subscribe
  • Alerts
  • Advertise
  • Contact
  • Conversations with Giants in Medicine
  • Author's Takes
  • Recently published
  • Brief Reports
  • Technical Advances
  • Commentaries
  • Editorials
  • Hindsight
  • Review series
  • Reviews
  • The Attending Physician
  • First Author Perspectives
  • Scientific Show Stoppers
  • Top read articles
  • Concise Communication
Top
  • View PDF
  • Download citation information
  • Send a letter
  • Terms of use
  • Standard abbreviations
  • Article usage
  • Citations to this article
  • Share this article
  • Need Help? E-mail the JCI
  • Top
  • Abstract
  • Supplemental material
  • Version history

Advertisement

ResearchIn-Press PreviewOphthalmology Free access | 10.1172/JCI131187

Deep learning predicts function of live retinal pigment epithelium from quantitative microscopy

Nicholas J. Schaub, Nathan A. Hotaling, Petre Manescu, Sarala Padi, Qin Wan, Ruchi Sharma, Aman George, Joe Chalfoun, Mylene Simon, Mohamed Ouladi, Carl G. Simon, Jr., Peter Bajcsy, and Kapil Bharti

Find articles by Schaub, N. in: JCI | PubMed | Google Scholar |

Find articles by Hotaling, N. in: JCI | PubMed | Google Scholar |

Find articles by Manescu, P. in: JCI | PubMed | Google Scholar

Find articles by Padi, S. in: JCI | PubMed | Google Scholar |

Find articles by Wan, Q. in: JCI | PubMed | Google Scholar

Find articles by Sharma, R. in: JCI | PubMed | Google Scholar |

Find articles by George, A. in: JCI | PubMed | Google Scholar

Find articles by Chalfoun, J. in: JCI | PubMed | Google Scholar

Find articles by Simon, M. in: JCI | PubMed | Google Scholar

Find articles by Ouladi, M. in: JCI | PubMed | Google Scholar

Find articles by Simon, Jr., C. in: JCI | PubMed | Google Scholar

Find articles by Bajcsy, P. in: JCI | PubMed | Google Scholar |

Find articles by Bharti, K. in: JCI | PubMed | Google Scholar

First published November 12, 2019 - More info

J Clin Invest. https://doi.org/10.1172/JCI131187.
Copyright © 2019, American Society for Clinical Investigation
First published November 12, 2019 - Version history
Abstract

Increases in the number of cell therapies in the preclinical and clinical phases have prompted the need for reliable and non-invasive assays to validate transplant function in clinical biomanufacturing. We developed a robust characterization methodology composed of quantitative bright-field absorbance microscopy (QBAM) and deep neural networks (DNNs) to non-invasively predict tissue function and cellular donor identity. The methodology was validated using clinical-grade induced pluripotent stem cell derived retinal pigment epithelial cells (iPSC-RPE). QBAM images of iPSC-RPE were used to train DNNs that predicted iPSC-RPE monolayer transepithelial resistance, predicted polarized vascular endothelial growth factor (VEGF) secretion, and matched iPSC-RPE monolayers to the stem cell donors. DNN predictions were supplemented with traditional machine learning algorithms that identified shape and texture features of single cells that were used to predict tissue function and iPSC donor identity. These results demonstrate non-invasive cell therapy characterization can be achieved with QBAM and machine learning.

Graphical Abstract
graphical abstract
Supplemental material

View

Version history
  • Version 1 (November 12, 2019): In-Press Preview

Article tools

  • View PDF
  • Download citation information
  • Send a letter
  • Terms of use
  • Standard abbreviations
  • Article usage
  • Citations to this article
  • Share this article
  • Need Help? E-mail the JCI

Go to:

  • Top
  • Abstract
  • Supplemental material
  • Version history
Advertisement
Advertisement
Follow JCI:
Copyright © 2019 American Society for Clinical Investigation
ISSN: 0021-9738 (print), 1558-8238 (online)

Sign up for email alerts