Related Experiment Video
Updated: May 21, 2026
08:28
Morphometric Protocol for the Objective Assessment of Blastocyst Behavior During Vitrification and Warming Steps
Published on: February 28, 2019
A method for semi-automatic grading of human blastocyst microscope images.
E Santos Filho1, J A Noble, M Poli
1Department of Engineering Science, Institute of Biomedical Engineering, University of Oxford, Oxford OX3 7DQ, UK. santfh@googlemail.com
Human Reproduction (Oxford, England)
|June 28, 2012
Summary
Automated grading of human blastocysts using computer image analysis is feasible. This method reduces subjectivity in embryo assessment, improving the precision of in vitro fertilization (IVF) treatments.
Area of Science:
- Reproductive biology
- Medical imaging
- Computational biology
Background:
- Accurate assessment of embryo viability is crucial for optimizing IVF success.
- Current embryo scoring systems rely on subjective visual analysis, leading to observer variability.
- Automating blastocyst grading can enhance objectivity and consistency in IVF procedures.
Purpose of the Study:
- To develop and validate a method for automated image segmentation and classification of human blastocysts.
- To reduce subjectivity and inter/intra-observer variation in embryo grading.
Main Methods:
- Advanced image analysis techniques (level set, phase congruency, ellipse fitting) for segmenting zona pellucida, trophectoderm (TE), and inner cell mass (ICM).
- Calculation of fractal dimension and texture descriptors (texture spectrum, grey-level run lengths) for TE and ICM.
- Support Vector Machine (SVM) classifiers for automatic grading based on morphological features.
Main Results:
- Fractal dimension of the TE boundary correlated well with cell number (r=0.81), a key indicator of blastocyst quality.
- Classifier accuracy ranged from 0.53 to 0.92, with the highest accuracy of 0.92 achieved for blastocyst grading.
- The system demonstrated varying accuracy levels for embryo development, ICM, and TE classification.
Conclusions:
- Semi-automatic computer-assisted grading of human blastocysts is a viable approach.
- This method offers a more precise and objective comparison of embryos.
- It can differentiate between embryos with similar morphological scores, potentially improving IVF outcomes.
