Can AI-guided feedback improve embryologists' selection of euploid embryos based on morphology alone?
Palmer et al., 2025
Selecting embryos with the correct number of chromosomes (euploid embryos) is a strategy that is sometimes employed to improve the chances of IVF success. However, genetic testing can be costly and is not always used or available in every IVF cycle.
This study investigated whether AI guidance could help embryologists better identify euploid embryos using only visual assessment. Embryologists were asked to rank embryos by their likelihood of being chromosomally normal, first using their expertise alone, then with AI-generated probability scores as guidance.
The AI system analyzed subtle visual features in embryo images that might indicate chromosome abnormalities. While embryologists' initial rankings showed some ability to identify euploid embryos, adding AI feedback significantly improved their accuracy.
Interestingly, the study found that embryologists sometimes overruled AI suggestions based on their experience, and these hybrid decisions often performed better than either human or AI alone. This demonstrates that AI works best as a decision-support tool rather than a replacement for human expertise. The findings suggest that AI-assisted embryo selection could help improve IVF outcomes, particularly in clinics without access to genetic testing.