Use of Artificial Intelligence Embryo Selection Based on Static Images to Predict First-Trimester Pregnancy Loss
Chavez-Badiola et al, 2024
This study looked at whether an artificial intelligence (AI) tool could help doctors predict the chance of early pregnancy loss (miscarriage) in couples who used in-vitro fertilization (IVF).
The AI program examined single snapshots of embryos taken before they were transferred into the womb. Based on these images, it grouped embryos into four quality categories: best, good, fair, and poor.
When the researchers checked how many patients suffered early miscarriages, they found that embryos rated as "best" or "good" were less likely to end in miscarriage than those rated as "fair" or "poor." Even when they looked only at embryos known to have a normal set of chromosomes (euploid), the trend was similar, although less clear.
Overall, the study suggests that this AI approach might help doctors and couples better understand an embryo’s chance of leading to a successful pregnancy. It’s an early step, and the researchers plan to do more studies with more patients, checking also the miscarriage tissue.
