HOW THE APPLICATION OF ARTIFICIAL INTELLIGENCE IS TRANSFORMING EMBRYO SELECTION IN IVF

Artificial intelligence (AI) is gradually being integrated into all areas of medicine, in the area of reproductive medicine AI is used to improve the selection and prediction of sperm, oocytes and embryos, as well as to generate better predictive models for in vitro fertilization (IVF). In Brazil, between 2020 and 2021, more than 36 thousand clinical pregnancies were registered with assisted human reproduction techniques, the use of AI in IVF treatments addresses questions such as whether such evaluations should be based on the performance of the classification or on the prediction of pregnancy. The present study performed an integrative literature review with the objective of describing the advances, impacts, and results of AI in assisted human reproduction, focusing on embryo selection in IVF. The PubMed, Lilacs and Google Scholar databases were used. The following descriptors were used: 'Embryo', 'artificial intelligence' and 'selection'. In all, 9 articles were selected published in the years 2020-2025. The results point out that it is not possible to define a set of 'universal' requirements and recommendations on how to divide datasets into training, validation, and testing sets, and that there is no universal truth about whether datasets should be balanced. Both precision medicine and AI suffer from a relative lack of standardization. The analysis of the different types of AI points out that the selection of embryos is crucial for primary intervention and prevention. There is a need for further clinical studies to confirm the applicability and safety of these models.