Evaluation of an Artificial Intelligence Model for the Prediction of Human Blastocyst Ploidy Without Invasive Procedures

Condition(s) studied

Investigational drug(s) / intervention(s)

Video recording: Videos of rotating the blastocysts will be recorded during the preparation stage of trophectoderm (TE) biopsy. The focal plane starts from the middle plane of the blastocyst. and then moves downwards until individual TE cells and inner cell mass (ICM) are clearly visible. A biopsy micropipette is used to gently push the blastocyst and rotate the blastocyst each time by a small angle, for instance, smaller than 35° such that more than 10 images can be captured for the entire 360° rotation to achieve high-accuracy measurement. After the first 360° rotation, the second one will be conducted around the axis perpendicular to the previous axis to ensure the whole surface of the blastocyst is captured. The entire rotation process will be video recoded.

Study summary

The goal of this clinical trial is to evaluate an artificial intelligence model for the prediction of human blastocyst ploidy without invasive procedures in couples that receive preimplantation genetic testing. The main questions it aims to answer are:

* Is an artificial intelligence model able to predict the ploidy status of a human blastocyst based on its 3D morphology?
* Do quantitative 3D morphological parameters of trophectoderm cells and inner cell mass have strong correlations with human blastocyst ploidy status?

Videos that include multi-view images of each blastocyst from participants will be collected on Day 5/6 of culture, and preimplantation genetic testing results of these blastocysts will be collected 4-8 weeks after trophectoderm biopsy.

Eligibility

Primary outcome measure(s)

• Sensitivity — Through study completion, an average of 1 year
  Sensitivity = TP/(TP+FN). TP: number of outcomes where a euploid blastocyst is correctly predicted as euploidy. FN: number of outcomes where a euploid blastocyst is incorrectly predicted as aneuploidy. High sensitivity indicates strong ability for the prediction of euploid blastocysts.
• Specificity — Through study completion, an average of 1 year
  Specificity = TN/(TN+FP). TN: number of outcomes where an aneuploid blastocyst is correctly predicted as aneuploidy. FP: number of outcomes where an aneuploid blastocyst is incorrectly predicted as euploidy. High specificity indicates strong ability for the prediction of aneuploid blastocysts.

Trial sites (5)

More The Affiliated Nanjing Drum Tower Hospital of Nanjing University Medical School trials in China