EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Using deep learning to predict the sex of human embryos

Mingshu Liang, Magdalena Zernicka-Goetz iD, Adiyant Lamba, Dilan Gokyer, Catherine King, Maciej Meglicki, María Luisa Pardiñas, María José de Los Santos, Changhuei Yang

Open Biology · 04.03.2026 · 16 (3)

DOI: 10.1098/rsob.250419 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Open Biology, online erschienen am 04.03.2026. Verfasst von Mingshu Liang und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical Engineering, California Institute of Technology, Pasadena, CA, USA. Hauptthemen laut MEDLINE: Deep Learning, Embryonic Development, Embryo, Mammalian, Sex Characteristics. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Cambridge Vice Chancellor's Award Fund, Medical Research Council (AL), Leverhulme Trust.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The existence of sex differences in human pre-implantation development remains an open question, with previous attempts based on human observations yielding inconclusive results. In this study, we combined manual annotation and deep learning analysis of a dataset comprising 515 time-lapse embryo movies to investigate whether birth sex influences early developmental dynamics. While manual assessment did not identify any developmental timing parameters that reliably distinguish male from female embryos, a deep learning model trained and tested on these videos achieves a statistically significant sex prediction accuracy of 61%. Importantly, our analyses identified the period after the eight-cell stage as critical for accurate prediction, indicating that subtle sex-related differences may begin to emerge around day 3 of human embryogenesis. Studying sex differences at this early stage may enhance our understanding of why some embryos fail to develop and why sex ratios can be skewed in the context of in vitro fertilization. More broadly, our findings open the possibility for an early, non-invasive detection tool that could assist in identifying and addressing sex-related embryonic developmental abnormalities.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningEmbryonic DevelopmentEmbryo, MammalianSex CharacteristicsSex Determination AnalysisHumansFemaleMaleFertilization in VitroTime-Lapse ImagingSex Ratio

Stichwörter der Autor:innen

in vitro fertilization (IVF)deep learningzygotic genes

Autor:innen und Institutionen

  • Department of Electrical Engineering, California Institute of Technology, Pasadena, CA, USA.
  • Gurdon Institute, University of Cambridge, Cambridge, UK.
  • Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge, UK.
  • Department of Innovation, Health Research Institute La Fe, Valencia, Spain.
  • Biophotonics Laboratory, Caltech, Pasadena, CA, USA.

ORCID-Profile: Magdalena Zernicka-Goetz

Förderung

  • Cambridge Vice Chancellor's Award Fund
  • Medical Research Council (AL)
  • Leverhulme Trust
  • Wellcome Trust
  • Heritage Research Institute for the Advancement of Medicine and Science at Caltech
  • Open Philanthropy

Zitieren

APA

Liang, M., Zernicka-Goetz, M., Lamba, A., Gokyer, D., King, C., Meglicki, M., Pardiñas, M. L., de Los Santos, M. J., & Yang, C. (2026). Using deep learning to predict the sex of human embryos. Open Biology, 16(3). https://doi.org/10.1098/rsob.250419

Vancouver

Liang M, Zernicka-Goetz M, Lamba A, Gokyer D, King C, Meglicki M, et al. Using deep learning to predict the sex of human embryos. Open Biology. 2026;16(3). doi:10.1098/rsob.250419

BibTeX
@article{Liang2026,
  author = {Liang, Mingshu and Zernicka-Goetz, Magdalena and Lamba, Adiyant and Gokyer, Dilan and King, Catherine and Meglicki, Maciej and Pardiñas, María Luisa and de Los Santos, María José and Yang, Changhuei},
  title = {Using deep learning to predict the sex of human embryos},
  journal = {Open Biology},
  year = {2026},
  volume = {16},
  number = {3},
  doi = {10.1098/rsob.250419}
}
Kostenlos starten →