Related Experiment Video
Updated: Jul 17, 2025
08:28
Morphometric Protocol for the Objective Assessment of Blastocyst Behavior During Vitrification and Warming Steps
Published on: February 28, 2019
8.8K
An artificial intelligence algorithm for automated blastocyst morphometric parameters demonstrates a positive
Yael Fruchter-Goldmeier1, Ben Kantor2, Assaf Ben-Meir2,3
1The Medical School for International Health and the Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel.
Scientific Reports
|September 5, 2023
Summary
Artificial intelligence enhances blastocyst selection for improved implantation potential. Larger blastocyst size, objectively measured, is significantly associated with higher success rates in embryo transfer.
Area of Science:
- Reproductive biology and assisted reproductive technology (ART).
- Computational biology and artificial intelligence in medicine.
Background:
- Current blastocyst selection relies on subjective morphological scoring and time-intensive morphokinetics.
- Need for objective, efficient, and accurate methods for blastocyst assessment.
Purpose of the Study:
- To evaluate the utility of an AI-driven semantic segmentation neural network for automated blastocyst morphometric analysis.
- To determine the association between automatically measured morphometric parameters and implantation potential.
Main Methods:
- A semantic segmentation neural network was developed to automatically measure blastocyst size, ICM size, ICM-to-blastocyst size ratio, and ICM shape.
- The model was trained on 1506 videos and validated on 102 videos.
- Univariable and multivariable logistic regression analyses were performed on data from 608 blastocyst transfers, adjusting for female age.
Main Results:
- Blastocyst size and ICM-to-blastocyst size ratio were significantly associated with implantation potential in univariable analysis.
- Multivariable analysis, adjusted for female age, confirmed blastocyst size as a significant predictor of implantation potential.
- Embryos with blastocyst size greater than the mean demonstrated 1.74 times higher odds of implantation. The algorithm achieved an AUC of 0.70.
Conclusions:
- Automated measurement of blastocyst morphometrics using AI is a precise, consistent, and time-saving tool.
- Larger blastocyst size is associated with increased implantation potential.
- AI-powered morphometric analysis can enhance blastocyst selection strategies in ART.
