Artificial Intelligence: Intervention for Infertility
Invitro Fertilization (IVF) treatment is expensive, exhausting, and overwhelming for infertile couples. Along with the time and money poured into IVF treatments couples’ risk not producing a viable pregnancy. Today, breakthrough research with artificial intelligence (A.I.) is improving the chances for couples who suffer from infertility. Cutting edge research in the field of A.I. show the capabilities of selecting worthwhile embryos, improving the chances for couples to successfully complete and sustain a pregnancy. Learning from the successes and failed attempts during IVF cycles, A.I. enhance the process by selecting the embryos whose score shows the highest chance to go on and become successful pregnancies.
A breakthrough in Australia, an A.I. named , has already proven her capabilities in selecting viable embryos for transfers. Using thousands of images and data provided from successful embryo transfers, IVY the A.I. predicted successful outcomes over 80% of the times.
In the near future, A.I. technologies will improve the chances for couple who are suffering with infertility to have a child using their own eggs. Despite the intense health regimen of women, poor egg quality consistently leads to poor embryo development. Without the aid of donor eggs, IVF doctored are likely to refuse future treatment after a few failed attempts without successful implantation from using the woman’s own eggs. However, the introduction of A.I. in the field of infertility is already showing auspicious results.
From Subjective to Objective Observations
A.I. technologies offer the promise of enhancing embryo selection. Most IVF clinics “grade” each embryo using scoring systems. However, most current embryo grading systems are subjective, based on the embryologist feelings and experiences. There are many other contributing factors involved that cannot be seen or measure with current technologies in use today. Generalizations about embryo quality thru the current grading systems are often inaccurate. There are many various ways in which an embryo grading system differs from clinic to clinic. To understand the concept of embryo grading in a nutshell, an embryo is multinucleated if more than one nucleus can be observed in any one cell. This typically develops on the 2 or 3 day. After day 3, it becomes extremely difficult for embryologists to identify the presence of multinucleation. Depending on the grade which will indicates the best or the worst embryo, this grading system is highly subjective and unique, depending on the expertise of any induvial embryologist.
Most if not all, multinucleated embryos are observed to be chromosomally abnormal. Embryo quality seen under the microscope in the IVF lab gives some reasonable ability to predict the chances for pregnancy soon after the transfer procedure. Predicting the outcomes of which embryo will result in a viable pregnancy can be greatly enhanced using and other A.I. technologies. Currently, Ivy can predict the potential of an embryo as far as the stage of a fetal heartbeat, a positive indicator that the transfer will result in a successful pregnancy.
Embryos from low-quality eggs often fail to develop properly. By objective analysis of the images, with no pre-determined restrictions, Ivy has taught itself to identify which embryos have the greatest likelihood of developing further. The embryo with the highest embryo score, and therefore the highest potential for leading to a viable fetus, can then be selected for transfer. Using A.I. in IVF treatment has the potential promise to fast-track the chances of a successful IVF transfer and in return, a healthy baby.
How AI will change the process of IVF
When eggs are removed from incubation for observation the risks of exposure are real. The difference in temperatures, PH differences and contamination, puts developing embryos at direct risk due to exposure. Using AI during these early stages can prevent excess exposure and ensure a higher quality embryo. In a typical IVF cycle, embryos are closely observed for a period of 3–5 days, before they are transferred. On the third day, good-quality embryos are expected to have 6- to 8-cells. Embryos that have not reached this stage within the first days of development, are not candidates for embryonic transfer.
As an artificial intelligence system, Ivy was able to review a massive amount of data, far more than any human could ever process. Hundreds of images from each embryo were reviewed across a large series of different embryos. Patterns were identified in the pictures which were then related back to whether each embryo developed into an ongoing pregnancy after the transfer procedure. Ivy was then put to the test through repetition. Using the results gathered, the data was compared against the outcomes of separate group of embryos. The results consistently showed which embryos were successful implants. Compared to any human embryology has achieved to date, these results alone, prove beyond doubt the importance of using an A.I. in embryo grading systems.
Better Outcomes for Couples from Implantation of A.I.
can improve the likelihood for infertile couples to conceive will be put the test this year with the public announcement of . The hope will be to increase the likelihood of infertile couples by objectively choosing the best suited embryos for implantation. Thousands of couples suffering from infertility will directly benefit from the use of this technology. The hope will be as more data is gathered to teach IVY what qualities make good eggs and embryos, better outcomes will be the result for couple who long sought to conceive equally, their own genetically related children.
Victoria Liset is strategic business & technology consultant to SMEs. She helps businesses improve their performance by using data more efficiently, and helping them to understand the implications of new technologies such as AI, Machine Learning, Big data, blockchain and IoT.