ASN's Mission

To create a world without kidney diseases, the ASN Alliance for Kidney Health elevates care by educating and informing, driving breakthroughs and innovation, and advocating for policies that create transformative changes in kidney medicine throughout the world.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005

email@asn-online.org

202-640-4660

The Latest on Twitter

Kidney Week

Abstract: FR-PO354

Deep Learning Predicts the Differentiation of Kidney Organoids Derived From Human-Induced Pluripotent Stem Cells

Session Information

Category: Development‚ Stem Cells‚ and Regenerative Medicine

  • 500 Development‚ Stem Cells‚ and Regenerative Medicine

Authors

  • Lee, Jong Young, Catholic University of Korea - Songeui Campus, Seoul, Korea (the Republic of)
  • Kim, Jinwon, Catholic University of Korea - Songeui Campus, Seoul, Korea (the Republic of)
  • Kim, Yong Kyun, Catholic University of Korea - Songeui Campus, Seoul, Korea (the Republic of)
  • Kim, Jin, Ajou University, Suwon, Gyeonggi-do, Korea (the Republic of)
  • Nam, Sun-ah, Catholic University of Korea - Songeui Campus, Seoul, Korea (the Republic of)
Background

Kidney organoids derived from human pluripotent stem cells (hPSCs) contain multi-lineage nephrogenic progenitor cells and can recapitulate the development of the kidney. Kidney organoids differentiated from human pluripotent stem cells can be applied in regenerative medicine as well as renal disease modeling, drug screening and nephrotoxicity testing. However, despite culturing under the same conditions, there are differences in the shape and growth level of each kidney organoid, making it difficult for clinical application. we hypothesized that an automated non-invasive method based on deep learning of bright-field images of kidney organoids can predict their differentiation status.

Methods

Kidney organoids were differentiated from induced pluripotent stem cells (iPSC). Bright-field images of kidney organoids were collected on day 18 after differentiation. To train the convolutional neural networks (CNNs), we utilized a transfer learning approach: CNNs were trained to predict the differentiation of kidney organoids on bright-field images, based on the mRNA expression of renal tubular epithelial cells as well as podocytes.

Results

The best-performing prediction model with DenseNet121 had a total Pearson correlation coefficient score of 0.783 on a test dataset. Furthermore, we focused on the classification of kidney organoids, into two categories: organoids with above-average gene expression (Positive) and those with below-average gene expression (Negative). Comparing the best-performing CNN with human-based classifiers, the CNN algorithm had a receiver operating characteristic-area under the curve (AUC) score of 0.85, while the experts had AUC score of 0.48. Time needed to classify one organoid by the experts took 1.04 seconds, but 0.014 seconds by CNN.

Conclusion

In this study, the CNN model predicts organoid maturity more accurately and rapidly than experts. These results show that the maturity of organoids can be predicted based on the bright-field and that the artificial intelligence algorithm can successfully recognize the differentiation status of kidney organoids.