Noninvasive methods for preimplantation blastocyst quality assessment in in vitro fertilization programs

Since the first in vitro fertilization (IVF) procedure, assisted reproductive technologies have helped many patients overcome infertility. However, according to the 2022 National Registry of Assisted Reproductive Technologies of the Russian Association of Human Reproduction, the probability of achieving pregnancy through IVF remains below 50%. Morphological assessment of blastocyst quality remains the gold standard. Implantation rates have increased to some extent due to the selection of high-quality embryos. However, given the subjectivity of morphological evaluation, further research is needed to establish the correlation between embryo reproductive potential and morphology. Time-lapse imaging combined with artificial intelligence may enhance the objectivity of assessment and identify additional morphological features indicative of blastocyst quality. The detection of exosomes, proteins, and metabolites secreted into the culture medium during embryo development may provide insights into the physiological state of the embryo and its interactions with the surrounding environment, potentially serving as markers of implantation potential. This review provides an overview of the morphological and biochemical markers of blastocyst quality, their interrelationships, and the use of artificial intelligence in embryo selection for transfer. A literature search was conducted in the electronic databases PubMed and Google Scholar using the following keywords: IVF, blastocyst, human embryo, culture media, timelapse system, embryo string, embryo exosomes, morphology, artificial intelligence, proteome, and metabolome. The analysis included studies published in the past five years.