During all these years of reproductive medicine, the values related to the morphology of the embryo were studied, which may be related to the possibilities of this implant. To this we must add that, in recent years, the study of the embryo has been carried out through time-lapse technology, allowing a more precise understanding of the development of the embryo, while serving to identify morphokinetic parameters as markers of the embryo viability, which makes it possible to define complementary models of embryo selection. However, after the vitrification and devitrification process, the blastocysts go through different changes in their morphology, which can make it difficult to assess their quality. In this sense, little is known to date about the application of this technology to vitrified and devitrified blastocysts.
Artificial Intelligence for embryo devitrification
Under this premise, the study “Analysis of the morphological dynamics of blastocysts after vitrification/warming: defining new predictive variables of implantation” was born. This study has been led by Dr. Mark Meseguerscientific supervisor of IVI and embryologist of IVI Valencia, presented today at the 10th International IVIRMA Congress.
“In our work we evaluate the post-devitrification dynamics of the embryos to predict the implantation potential of devitrified blastocysts through the use of Artificial Neural Networks (RNA) based on Artificial Intelligence (AI). In this sense, we are working on an AI algorithm that studies the behavior of the embryo from the time it is devitrified until it is transferred, which lasts approximately 4 hours. Thus, AI shows us that an embryo that begins its expansion early (when the average expansion time is 50 minutes) and carries out this process quickly, acquiring a surface greater than 0.14 square millimeters, can reach to implant up to 30% more than an embryo that expands later and slowly during those first 4 hours of life. AI thus allows us to identify embryos that, although they show good morphology, have a low probability of implantation because when they devitrify, they have either taken a long time to expand or have expanded very little.”explains Dr. Marcos Meseguer.
What does the study consist of?
The study led by Dr. Meseguer comprises a retrospective analysis on a sample of 511 devitrified blastocysts, with the main objective of describing the parameters involved in the morphology of blastocysts that have been vitrified and then devitrified, during the time that elapses. between this process and embryo transfer, in an attempt to better understand the embryo expansion procedure.
“When we vitrify the embryo, we leave it in an inert state, removing the water that is what moves all the machinery of the cell. The moment you remove the water, it is as if time stood still and the embryo can remain like this for years, without time having a negative impact on its quality. When we reactivate time, we put the water back into the embryo, which enters little by little and does not do so in the same way in all embryos. This process of entering water and leaving the antifreeze -which is the cryoprotectant- is not carried out by all embryos in the same way, nor do they all start at the same time. And this is the starting point of our work: we have seen that the embryo in which water begins to enter earlier presents a better prognosis. And the embryo that expands more quickly is going to do better than the one that expands more slowly. This leads us to correlate the re-expansion of devitrified blastocysts with their chances of implantation. Thus, more than 60% of the re-expanded blastocysts implanted successfully, compared to 6% of those that did not re-expand after devitrification”, comments Dr. Meseguer.
Artificial Intelligence to improve reproductive success
Today both prolonged embryo culture and blastocyst phase transfer are common. Both things have shown an improvement in the selection of embryos and, therefore, in the success rates achieved with assisted reproduction treatments. This strategy assumes that all viable blastocysts are cryopreserved and transferred in subsequent cycles, also avoiding the risk of ovarian hyperstimulation.
This is what is known as deferred transfer cycles and its increase implies a further development of more and more precise selection criteria, allowing the improvement of the results of the transferred vitrified embryos.
“It is well known that each observation involves exposure to suboptimal conditions outside the controlled environment of an incubator, which can potentially affect treatment success. Hence, continuous monitoring of devitrified blastocysts using time-lapse systems can provide us with valuable information about their implantation potential while they remain in a stable and controlled culture environment. At this point, it is important to emphasize that all blastocysts were vitrified and devitrified using the Cryotop method, and were placed in the Embryoscope immediately after devitrification until transfer. In addition, another differential element of our work points to the fact that it provides objective quantitative values for the variables involved in blastocyst re-expansion, unlike the subjective morphological evaluation that has been used for blastocyst re-expansion performed to date.” concludes Dr. Meseguer.
All in all, it can be concluded that making use of Artificial Intelligence to analyze the dynamics of vitrified and devitrified blastocysts could be useful to predict their implantation potential. And, therefore, the predictive models in vitrified cycles can assume that vitrified embryos are not transferred with a low success rate. However, the observed correlations and the proposed algorithm need to be validated in a prospective trial to assess its efficacy.
