A Caltech Library Service

Stain-free detection of embryo polarization using deep learning

Shen, Cheng and Lamba, Adiyant and Zhu, Meng and Zhang, Ray and Zernicka-Goetz, Magdalena and Yang, Changhuei (2022) Stain-free detection of embryo polarization using deep learning. Scientific Reports, 12 . Art. No. 2404. ISSN 2045-2322. doi:10.1038/s41598-022-05990-6.

[img] PDF - Published Version
Creative Commons Attribution.

[img] PDF - Submitted Version
Creative Commons Attribution Non-commercial No Derivatives.

[img] MS Word - Supplemental Material
Creative Commons Attribution.


Use this Persistent URL to link to this item:


Polarization of the mammalian embryo at the right developmental time is critical for its development to term and would be valuable in assessing the potential of human embryos. However, tracking polarization requires invasive fluorescence staining, impermissible in the in vitro fertilization clinic. Here, we report the use of artificial intelligence to detect polarization from unstained time-lapse movies of mouse embryos. We assembled a dataset of bright-field movie frames from 8-cell-stage embryos, side-by-side with corresponding images of fluorescent markers of cell polarization. We then used an ensemble learning model to detect whether any bright-field frame showed an embryo before or after onset of polarization. Our resulting model has an accuracy of 85% for detecting polarization, significantly outperforming human volunteers trained on the same data (61% accuracy). We discovered that our self-learning model focuses upon the angle between cells as one known cue for compaction, which precedes polarization, but it outperforms the use of this cue alone. By compressing three-dimensional time-lapsed image data into two-dimensions, we are able to reduce data to an easily manageable size for deep learning processing. In conclusion, we describe a method for detecting a key developmental feature of embryo development that avoids clinically impermissible fluorescence staining.

Item Type:Article
Related URLs:
URLURL TypeDescription Paper ItemData/Code
Shen, Cheng0000-0001-7136-4715
Zhu, Meng0000-0001-6157-8840
Zernicka-Goetz, Magdalena0000-0002-7004-2471
Yang, Changhuei0000-0001-8791-0354
Additional Information:© The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit Received 22 October 2021; Accepted 10 January 2022; Published 14 February 2022. We thank all colleagues in the C.Y. and M.Z.-G. labs for helpful suggestions and feedback. We also thank all human volunteers. We thank all funding sources: Wellcome Trust (098287/Z/12/Z) (MZG), Leverhulme Trust (RPG- 2018-085) (MZG), Open Philanthropy/Silicon Valley (MZG), Weston Havens Foundations (MZG), NIH R01 HD100456-01A1 (MZG), Rosen Bioengineering Center Pilot Research Grant Award (9900050) (CY, MZG), Medical Research Council (AL), Cambridge Vice Chancellor’s Award Fund (AL). Data availability: The testing dataset is available on for model validation use and academic purposes only. All other datasets generated and analyzed in the current study (including larger training image dataset) are available from the corresponding author (M.Z.-G) on reasonable request. Code availability: The training code for the single DCNNs and the testing code for the ensemble DL model are available at: Author Contributions: These authors contributed equally: Cheng Shen and Adiyant Lamba. C.S. and A.L. were responsible for planning project directions, interpretation of results and optimization of the model. C.S. was responsible for the design of the model. M.Z. and A.L. were responsible for embryo recordings and assembly of the dataset. A.L. was responsible for annotating embryo images. The project was conceived by M.Z., M.Z.-G and C.Y. and supervised by M.Z.-G and C.Y. The manuscript was written by A.L., C.S., C.Y. and M.Z.-G. R.Z. edited the manuscript. The authors declare no competing interests.
Group:Rosen Bioengineering Center
Funding AgencyGrant Number
Wellcome Trust098287/Z/12/Z
Leverhulme TrustRPG-2018-085
Open PhilanthropyUNSPECIFIED
Weston Havens FoundationUNSPECIFIED
NIHR01 HD100456-01A1
Donna and Benjamin M. Rosen Bioengineering Center9900050
Medical Research Council (UK)UNSPECIFIED
University of CambridgeUNSPECIFIED
Subject Keywords:Computational science; Developmental biology; Embryogenesis; Embryology
Record Number:CaltechAUTHORS:20220215-496417800
Persistent URL:
Official Citation:Shen, C., Lamba, A., Zhu, M. et al. Stain-free detection of embryo polarization using deep learning. Sci Rep 12, 2404 (2022).
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:113452
Deposited By: Tony Diaz
Deposited On:15 Feb 2022 21:23
Last Modified:15 Feb 2022 21:23

Repository Staff Only: item control page