Eugin presents a study in which an artificial intelligence model was developed to prescribe medication with the same precision and reliability as a doctor

  • Paper presented at the 37th Annual Congress of the European Society of Human Reproduction and Embryology (ESHRE)
  • The model, developed in partnership with the CSIC’s Artificial Intelligence Research Institute, is based on data collected from a sample of 2,713 patients, and is designed to identify optimal drug doses for ovarian stimulation in assisted reproduction procedures
  • Another study presented by Eugin at ESHRE, involving a sample of 14,000 embryos, concluded that slow embryos are more likely to be abnormal and not thrive, which increases with the woman’s age

Researchers at Eugin, the assisted reproduction and fertility group, have developed, in partnership with the CSIC Artificial Intelligence Research Institute, an artificial intelligence algorithm that can identify the optimal dose of medication for ovarian stimulation required by a patient in an in vitro fertilisation cycle. The model, which was developed through machine learning, prescribes as accurately and reliably as a doctor.

The algorithm it uses is based on data collected from a sample of 2,713 patients and its functional performance was validated with a further 524 patients, where the model’s prescription was then validated based on the specialists’ experience.

“This system can be used in different ways in clinical practice ranging from a support tool for physicians who are new to the area or quality control for the more experienced ones, to a second medical opinion,” said Núria Correa, the study’s lead researcher and a participant in the industrial doctorate programme promoted by the Generalitat de Catalunya (the Catalan government).

Eugin, who presented this paper at the 37th Annual Congress of the European Society of Human Reproduction and Embryology (ESHRE), has been working for years on artificial intelligence as part of two of its main objectives: the application of medicine based on scientific evidence and a commitment to research. “With the application of artificial intelligence techniques to assisted reproduction, we obtain a greater scientific yield from the data, which enables us to implement a kind of personalised medicine based on their particular cases,” Correa emphasised.

International research with large samples

During the 2021 ESHRE, Eugin Group presented a total of 10 studies on different topics in the field of fertility and assisted reproduction. Research is one of the company’s key areas and, in this regard, the group’s participation in the congress was also notable this year for two studies that were carried out thanks to the extensive network of clinics they have all over the world and in which those based in Spain, Brazil and the United States took part.

The first study reviewed a sample of 11,000 embryos that underwent genetic testing (PGT-A), which makes it possible to identify chromosomal abnormalities and conclude whether an embryo is suitable for transfer. Using this test, it was found that, in women aged 37 years on average, the proportion of abnormal embryos does not change, even if the number of embryos obtained increases.

Meanwhile, the second study analysed a sample of 14,000 embryos that also underwent PGT-A in order to determine whether slower developing embryos (those that reach the optimal stage for this test later) were more likely to be abnormal than embryos that develop at normal speed (on day 5). The study concluded that slow embryos are more likely to be abnormal and, therefore, not thrive, which is something that increases as the woman ages.


View the research poster “Development and validation of an Artificial Intelligence algorithm that matches a clinician ability to select the best follitropin dose for ovarian stimulation”.

View the research poster “Blastocyst cohort size is not associated with embryo aneuploidy: comprehensive multi centre data from current preimplantation genetic testing cycles”

View the research poster “Delayed blastocyst development is associated with a higher risk of aneuploidy in patients of advanced maternal age”

Share this: