Related Experiment Video
Updated: Nov 26, 2025
08:28
Morphometric Protocol for the Objective Assessment of Blastocyst Behavior During Vitrification and Warming Steps
Published on: February 28, 2019
9.1K
Automated Measurements of Key Morphological Features of Human Embryos for IVF
B D Leahy1,2, W-D Jang1, H Y Yang3
1School of Engineering and Applied Sciences,Harvard University, Cambridge MA 02138, USA.
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|December 14, 2020
Summary
Automating human embryo analysis using machine learning speeds up the selection of the highest quality embryos for In-Vitro Fertilization (IVF). This AI-powered approach enhances embryo assessment for better pregnancy outcomes.
Area of Science:
- Reproductive biology
- Medical imaging
- Artificial intelligence
Background:
- Selecting viable embryos is crucial for successful In-Vitro Fertilization (IVF) outcomes.
- Manual analysis of time-lapse microscopy videos for embryo assessment is time-consuming and subjective.
- Automated analysis can improve efficiency and objectivity in embryo selection.
Purpose of the Study:
- To develop and validate a machine learning pipeline for automated feature extraction from time-lapse microscopy of human embryos.
- To enhance the quantitative assessment of embryo quality for improved In-Vitro Fertilization (IVF) success rates.
Main Methods:
- A pipeline of five convolutional neural networks (CNNs) was developed for automated analysis.
- The pipeline includes semantic segmentation of embryo regions, fragment severity regression, developmental stage classification, and cell/pronuclei instance segmentation.
- This method processes time-lapse microscopy data of human embryos.
Main Results:
- The machine learning pipeline successfully automates feature extraction from time-lapse embryo microscopy.
- Quantitative, biologically relevant features can be measured rapidly.
- The approach significantly reduces the time and subjectivity associated with manual embryo analysis.
Conclusions:
- Automated feature extraction using CNNs offers a promising solution for objective and efficient human embryo assessment in In-Vitro Fertilization (IVF).
- This technology has the potential to aid clinicians in selecting the highest quality embryos, thereby improving pregnancy rates.
