• Application of artificial interlligence in embryo selection
  • Hamidreza Sadeghsalehi,1,*
    1. Department of Neuroscience, Faculty of Advanced Technologies in Medicine, Iran University Of Medical Sciences,


  • Introduction: As fertility rates have declined in recent years, assisted reproduction techniques (ART) have been more widely used. In vitro fertilization (IVF) results are influenced by many variables and their interwoven relationships. Embryo quality is unquestionably a crucial determinant of a successful IVF outcome. Artificial intelligence (AI) has been widely employed to enhance and automate embryo selection in recent years. We discuss the latest studies on AI in embryo selection in this review.
  • Methods: The current study employed a review method that searched for studies published in PubMed, Web of Science, and Google Scholar up to April 2022. Without consideration for language, databases were searched for relevant keywords. All relevant studies are included in this study.
  • Results: Several models have been developed, and while some of them have shown potential, there are still numerous obstacles to overcome. Some models provide a single still image of the embryo, while others have a time-lapse video of the embryo's development. Female age, number of previous treatments, stimulation procedure, clinic-specific parameters, and manual annotations of morphological and kinetic data could all be included in other samples. Including such entries can dramatically improve performance metrics. The majority of studies used convolutional neural network (CNN) to develop their models. There were considerable differences in model optimization between the studies. Because of the wide range of input, embryo population, and outcome, it's difficult, if not impossible, to compare AI performance outcomes across studies.
  • Conclusion: According to the findings of our study, AI has a lot of potential for the future of embryo selection; nevertheless, they still have a long way to go before they can claim to be as good as clinical embryologists in predicting outcomes.
  • Keywords: Artificial intelligence, embryo selection, In vitro fertilization