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

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

doi:10.1172/jci131187

Cited by 57 publications

(59 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The quality of generated RPE sheets also can vary among different hiPSC lines 13 and among different protocols in the same hiPSC line 57 . Consistent with previous reports 19,23 , we also found that TER values varied between RPE sheets even using the same hiPSC cell line and the same culture protocol. For industrialization of cellular products for regenerative medicine, it must be ensured that every product is fully functional.…”

Section: Discussionsupporting

confidence: 92%

“…Replating RPE6iN-induced RPE cells onto transwell produced mature RPE sheets at high purity without manual selection. We found lot-to-lot variations of barrier function between mature RPE sheets, consistent with general issues reported in previous studies 19,23 . To support the robust RPE sheet production ability, we developed a machine learning-based prediction model to predict failure samples only from their non-labeled cellular morphologies in microscopic images.…”

supporting

confidence: 92%

“…Microscopic image-based analyses have been expected to assist in the evaluation of cell products 18 . Functional prediction models based on microscopic observation of the degree of pigmentation in RPE cells have been reported 19,20 . The morphological features of cells contain additional information about the cellular status and have been conventionally used in the daily monitoring of cells.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Reproducible production and image-based quality evaluation of retinal pigment epithelium sheets from human induced pluripotent stem cells

Yao

Takemoto

Ito

et al. 2020

Sci Rep

View full text Add to dashboard Cite

transplantation of retinal pigment epithelial (Rpe) sheets derived from human induced pluripotent cells (hipSc) is a promising cell therapy for Rpe degeneration, such as in age-related macular degeneration. current Rpe replacement therapies, however, face major challenges. they require a tedious manual process of selecting differentiated RPE from hiPSC-derived cells, and despite wide variation in quality of RPE sheets, there exists no efficient process for distinguishing functional RPE sheets from those unsuitable for transplantation. to overcome these issues, we developed methods for the generation of Rpe sheets from hipSc, and image-based evaluation. We found that stepwise treatment with six signaling pathway inhibitors along with nicotinamide increased RPE differentiation efficiency (RPE6iN), enabling the RPE sheet generation at high purity without manual selection. Machine learning models were developed based on cellular morphological features of f-actin-labeled Rpe images for predicting transepithelial electrical resistance values, an indicator of Rpe sheet function. Our model was effective at identifying low-quality RPE sheets for elimination, even when using label-free images. The RPE6iN-based RPE sheet generation combined with the non-destructive image-based prediction offers a comprehensive new solution for the large-scale production of pure Rpe sheets with lot-to-lot variations and should facilitate the further development of Rpe replacement therapies. Human pluripotent stem cells (hPSC) [human embryonic stem cells (hESC) and human induced pluripotent stem cells (hiPSC)] are valuable sources for regenerative therapies because of their capacity for self-renewal and pluripotency 1. Retinal regeneration using hPSC is a new promising treatment at the cutting edge of stem cellbased regeneration therapies. Retinal pigment epithelium (RPE) cells play important roles in vision and retinal function maintenance. Transplantation of RPE suspension or RPE sheets derived from hPSC for age-related macular degeneration (AMD) and Stargardt disease patients is considered safe and potentially effective 2-5. However, this treatment faces challenges regarding the purity and quality control of RPE products for transplantation therapy. The properties of cellular products are totally different from those of chemical drugs that are wellestablished in the synthesis and quality control of drug development.

show abstract

Section: Discussionsupporting

confidence: 92%

supporting

confidence: 92%

mentioning

confidence: 99%

See 1 more Smart Citation

Reproducible production and image-based quality evaluation of retinal pigment epithelium sheets from human induced pluripotent stem cells

Yao

Takemoto

Ito

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In addition to iPS cell line validation, functional analyses of the patient iPSC-RPE in quantitative assays examining drusen-like particles, autophagic pathways, inflammatory cytokine secretion, shape, pigmentation, and other structural features will also enable a comprehensive analysis of RPE physiology paired with genotype information. 59 NYSCF has completed the first ten cell lines, and some of the isogenic control lines (2/10) have been completed and are in the process of being validated. By statute, these iPSC lines cannot be directly used in humans or for any diagnostic, prognostic, or treatment purposes.…”

Section: Launching a Resource For The Amd Research Communitymentioning

confidence: 99%

NEI-Supported Age-Related Macular Degeneration Research: Past, Present, and Future

Wright

Mazzucco

Becker

et al. 2020

Trans. Vis. Sci. Tech.

View full text Add to dashboard Cite

To review past and current National Eye Institute (NEI)-supported age-related macular degeneration (AMD) activities and initiatives and preview upcoming coordinated efforts for studying AMD. Methods: We conducted and summarized a portfolio analysis and literature review of NEI intramural and extramural AMD activities. Results: The NEI supports a broad range of AMD research, both by individual independent investigators as well as through networks and consortia. The International AMD Genomics Consortium, Age-Related Eye Disease Study, Age-Related Eye Disease Study 2 (AREDS2), and Comparison of AMD Treatments Trial legacy work probed the complex genetics, clinical presentation, and standards of patient care, respectively. The NEI AMD Pathobiology Working Group identified gaps and opportunities for future research efforts. The AMD Ryan Initiative Study and clinical trials testing the efficacies of minocycline to modulate retinal microglia activity and induced pluripotent stem cells-derived retinal pigmented epithelium (RPE) patch implants to rescue photoreceptor cell death are among the future directions for NEI-supported AMD research. Finally, NEI commissioned the creation of AREDS2 participant-derived induced pluripotent stem cell (iPSC) lines linked to their associated genomic and phenotypic datasets. These datasets will also be linked to the data obtained using their associated iPSC-derived cells (RPE, retina, choroid) and made publicly available. Conclusions: Investments by NEI for AMD research will continue to provide invaluable resources to investigators committed to addressing this complex blinding disease and other retinal degenerative diseases.

show abstract

“…This approach has found a lot of applications in multiple fields of biology and medicine. For example, for diagnosis of diabetic retinopathy based on fundus imaging (Gulshan et al, 2016 ) and for skin cancer classification (Esteva et al, 2017 ), and recently it was proven effective to predict the very early onset of PSC differentiation (Waisman et al, 2019 ) and the quality of retinal pigment epithelium (RPE) differentiation in a two-dimensional setting (Schaub et al, 2020 ). Being inspired by the success that this approach showed on the prediction of spontaneous differentiation of PSCs with basic bright-field imaging used as a source of information, we hypothesized that basic-contrast bright-field images contain sufficient information on tissue specification, and it is possible to extract it using convolutional neural networks.…”

Section: Introductionmentioning

confidence: 99%

Convolutional Neural Networks Can Predict Retinal Differentiation in Retinal Organoids

Kegeles

Наумов

Karpulevich

et al. 2020

Front. Cell. Neurosci.

View full text Add to dashboard Cite

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

Cited by 57 publications

References 34 publications

Reproducible production and image-based quality evaluation of retinal pigment epithelium sheets from human induced pluripotent stem cells

Reproducible production and image-based quality evaluation of retinal pigment epithelium sheets from human induced pluripotent stem cells

NEI-Supported Age-Related Macular Degeneration Research: Past, Present, and Future

Convolutional Neural Networks Can Predict Retinal Differentiation in Retinal Organoids

Contact Info

Product

Resources

About