Embryo Ranking Intelligent Classification Algorithm (ERICA): artificial intelligence clinical assistant predicting embryo ploidy and implantation
Chavez-Badiola et al., 2020
Selecting the best embryo for transfer is crucial for successful IVF outcomes. Currently, embryologists assess embryos by eye under a microscope, evaluating features like expansion and cell appearance. Alternatively, clinics may use time-lapse imagery or genetic testing (PGT-A), which requires removing cells from the embryo.
This study introduces ERICA, an AI-powered algorithm that analyzes single static images of embryos. Using deep learning, ERICA extracts 94 visual features and combines these with patient age and embryo development time to predict which embryos are most likely to be genetically normal and implant successfully.
This study found that ERICA correctly predicted genetic normalcy with 70% accuracy and placed a genetically normal embryo at the very top of its ranking in 79% of cases, outperforming random selection and two senior embryologists working from the same images.
This technology could help embryologists make more informed embryo selection decisions without requiring costly time-lapse equipment or removing cells from the embryo, improving embryo prioritization and potentially reducing time to pregnancy.