The University of Auckland
Browse

Updated labels for open source embryo morphokinetic dataset.

Download (30.83 MB)

CSV file contains a new label (empty) for an open source embryo morphokinetic dataset[1]. Updating these labels was done with a EfficientNetV2 Large built upon a proprietary dataset. Relabeling these images had an outsized effect on model performance in our work[2].

[1] Gomez T, Feyeux M, Boulant J, Normand N, David L, Paul-Gilloteaux P, Fréour T, Mouchère H. A time-lapse embryo dataset for morphokinetic parameter prediction. Data Brief. 2022 May 11;42:108258. doi: 10.1016/j.dib.2022.108258. PMID: 35599827; PMCID: PMC9120221.

[2] Misaghi et. al. Accurate Machine Learning Model for Human Embryo Morphokinetic Stage Detection

Funding

Improving IVF success rates through machine learning

Ministry of Business, Innovation and Employment

Find out more...

History

Publisher

University of Auckland

Usage metrics

    FMHS

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC