Related Experiment Video
Updated: May 16, 2026
09:03
Semiconductor Sequencing for Preimplantation Genetic Testing for Aneuploidy
Published on: August 25, 2019
Improving embryo ploidy prediction: a machine learning approach using morphokinetic meta-variables and clinical data.
Enric Güell-Penas1, Minerva Ferrer-Buitrago2, Empar Ferrer I Robles2
1Consultfiv Data Science, Valls, Spain; Centre Procrear, Reus, Spain.
Reproductive Biomedicine Online
|May 14, 2026
Summary
A new machine learning model, LIFE Predict v1.1, accurately predicts embryo aneuploidy using time-lapse morphokinetic data and clinical information. This tool enhances non-invasive embryo selection by providing risk stratification beyond traditional morphological assessment.
Area of Science:
- Reproductive Medicine
- Artificial Intelligence in Healthcare
- Embryology
Background:
- Non-invasive embryo selection is crucial for improving in vitro fertilization (IVF) success rates.
- Conventional morphological assessment has limitations in accurately predicting embryo ploidy status.
- Time-lapse imaging offers dynamic insights into early embryo development.
Purpose of the Study:
- To develop and validate a machine learning model integrating time-lapse morphokinetic meta-variables with clinical data for accurate prediction of embryo aneuploidy.
- To assess the model's performance in non-invasive embryo selection for improved IVF outcomes.
Main Methods:
- A retrospective multicentre cohort study involving 1190 blastocysts from nine fertility clinics.
- Development of the LIFE Predict v1.1 model using time-lapse morphokinetic meta-variables (range and mean absolute error) and clinical data.
- Model validation using area under the receiver operating characteristic curve (AUC-ROC) and confusion matrix metrics, with logistic regression for odds ratio (OR) calculation.
Main Results:
- The LIFE Predict v1.1 model achieved high accuracy with an AUC of 0.824 in cross-validation and 0.818 in external validation.
- A significant inverse relationship was found between the LIFE Predict v1.1 score and aneuploidy risk (OR = 1.76 per one-point decrease).
- Aneuploidy rates decreased significantly across ascending score quartiles, demonstrating effective risk stratification, even within conventionally graded embryos.
Conclusions:
- The LIFE Predict v1.1 model accurately predicts embryo aneuploidy using morphokinetic meta-variables and clinical data.
- This model offers actionable risk stratification, complementing morphological assessment for enhanced non-invasive embryo selection.
- Prospective clinical validation is recommended to confirm the real-world utility of the LIFE Predict v1.1 model.
