Picture Perfect?: Determining the Clinical Utilization of Artificial Intelligence in Oocyte Cryopreservation

Abstract

The two different culture media employed in this study had a significant (p<0.01) impact both on the dynamics of post-fertilization development and embryo quality, as assessed by subjective morphological parameters. Media-A had a significantly higher proportion of embryos reaching the blastocyst stage at <110 hours and a higher proportion of top-quality blastocysts. The area under the receiver operating curve obtained from the ML models was 80.9% for Media-A and 80.5% for Media-B. The models were deployed as a web-based service running a Python script to extract the relevant morphokinetic and clinical parameters of embryos in real-time to provide the predicted probability of 110-hour blast formation.CONCLUSIONS: There were significant differences in the rate of embryo development and the proportion of top-graded blastocysts between the two culture media. The models developed to estimate the probability of reaching the blastocyst stage in <110 hours had a similar ROC-AUC but different structures to account for the different dynamics of development induced by the culture media.IMPACT STATEMENT: This work showcases the importance of continuously evaluating the characteristics of embryo development as the IVF laboratory evolves. Models implemented for embryo evaluation need to be adjusted accordingly. This study also suggests that embryo development is significantly impacted by laboratory-specific parameters; therefore, ML models should be tailored to the ART setting in which they will be utilized. Finally, deploying these models in platforms that allow for rapid and seamless updates to new versions increases the ability of these decision-support systems to adapt to the continuously evolving IVF laboratory.SUPPORT: Institutional

Innovations for Optimizing Fertility and Ovarian Functionality in Female Cancer Patients

“…Recently, artificial intelligence has been applied to determine egg quality for women undergoing oocyte cryopreservation. 36 Violet technology can predict the probabilities of blastocyst development and live birth according to the quality of each egg, through application of a light microscope to capture images of the oocytes before they are frozen, and using a database comparison system. Automated closed-vitrification systems are also constantly being improved, aiming to avoid decreases in the efficacy of the vitrification protocol.…”

Data-Driven Medicine in the Diagnosis and Treatment of Infertility

“…AI can be applied to analysing oocyte images to identify oocytes with the best reproductive potential (competent oocytes), something that a trained embryologist cannot do at present [ 110 ]. AI methods could not only improve the ART performance, but also provide an accurate evaluation of oocyte vitrification and warming procedures [ 115 ], which is of particular interest, as oocyte cryopreservation is becoming increasingly popular. Additionally, by using non-invasive AI methods, the quality of the ovarian yield could be analysed retrospectively after an ovarian stimulation cycle, which may be helpful in managing psychosocial expectations of intended parents.…”