In the world of assisted reproduction, one of the major concerns of professionals and patients is selecting the most viable embryo for a successful pregnancy. For years, embryologists and scientists have sought methods to identify ‘healthy’ embryos, free of genetic alterations that could complicate the process. Now, with the advent of artificial intelligence (AI), embryo selection is taking a turn towards a more accurate, non-invasive and effective approach.
This article explores how AI is revolutionising embryo selection, improving success rates in assisted reproductive treatments and offering hope to many people struggling to have a child.
The challenge of Embryo Selection
Each assisted reproductive technology (IVF) cycle may result in the generation of several embryos, but not all of them have the same quality or likelihood of success. The goal in IVF treatments is to identify the embryo that has the best chance of successfully implanting in the mother’s uterus and eventually resulting in a ‘healthy’ pregnancy.
However, many embryos have chromosomal abnormalities, such as aneuploidy (an incorrect number of chromosomes), which can result in miscarriage, implantation failure or genetic disease. This is one of the main causes of IVF failures, which creates a significant emotional burden for people trying to conceive.
In addition, some systems also incorporate free DNA (cfDNA) data that embryos release into the environment in which they develop. This data allows AI to obtain genetic information without the need for invasive biopsy. Recent studies have shown that cfDNA analysis, combined with artificial intelligence, can help to more accurately predict which embryos are genetically viable for transfer (Treff, et al., 2020).
Benefits of AI in Embryo Selection
Best Success Rates
One of the most notable advantages of AI in embryo selection is its ability to increase pregnancy rates. AI can accurately predict which embryos are most likely to implant, resulting in a higher success rate in IVF treatments. By selecting the most viable embryos, the likelihood of transferring aneuploid embryos is reduced, which in turn decreases miscarriage and implantation failure rates (Morbeck, et al., 2019).
Non-invasive approach
AI-based technology allows embryos to be selected without the need for invasive biopsies. Instead of removing cells from the embryo for genetic analysis, AI uses images and data from free DNA found in the developing environment. This not only makes the process less invasive for the embryo, but also reduces the risk of potential damage, increasing the overall safety of the treatment (Rienzi, et al., 2017).
Increased Accessibility and Reduced Costs
AI analysis can be cheaper and more accessible than traditional biopsies. As AI algorithms improve and become more standardised, the costs associated with their implementation in fertility clinics can be reduced, making this technology more accessible to a wider range of embryology professionals. In addition, by selecting the right embryo from the start, additional treatment cycles that might be necessary when an embryo fails to implant successfully are reduced (Salumets, et al., 2019).
Time Optimisation
AI algorithms can analyse large amounts of data in minutes or even seconds, speeding up the embryo selection process. This allows clinicians and embryologists to focus on other aspects of treatment, without wasting valuable time on difficult decisions or manual image review (Zhang, et al., 2020).
Success Stories and Recent Developments
The use of artificial intelligence in embryo selection is not just a theory; real, measurable results are already being obtained. In a study by Modell, et al. (2021), it was shown that AI can predict with more than 80% accuracy which embryos would be euploid (i.e. with the correct number of chromosomes). This level of accuracy is comparable to traditional biopsy methods, but without the associated risks.
In addition, platforms such as Life Whisperer Genetics and CHLOE EQ are implementing AI in fertility clinics to improve embryo selection. These platforms have demonstrated that AI can be as effective as traditional methods, but with greater efficiency and less invasiveness (Pauli, et al., 2020).
Ethical Challenges and Considerations
Despite promising advances, the use of AI in embryo selection raises certain challenges and ethical questions that need to be addressed. First, the accuracy of the algorithms is critical. Although AI can identify patterns, it still depends on the quality of the data used to train the algorithms. This means that if the data is not sufficiently representative or accurate, the results may not be reliable.
In addition, there are ethical concerns about how this technology will be used in the future. While AI is a powerful tool for detecting genetic diseases, some fear that it could be used to select embryos with non-health-related characteristics, opening up the debate about ‘eugenics’ and the ethical limits of genetic manipulation (Caplan, et al., 2020).
It is essential that clear regulations are established on the use of AI in embryo selection, ensuring that it is used for well-defined medical and ethical purposes, and that it is not diverted towards the selection of non-medical characteristics.
The Future of AI Assisted Reproduction
The future of in vitro fertilisation and embryo selection is clearly marked by artificial intelligence. Over time, AI algorithms are expected to continue to improve, allowing for greater accuracy in embryo selection and thus higher treatment success rates. In addition, the technology will continue to evolve to offer more accessible and cost-effective solutions for people facing fertility problems.
Conclusion
Artificial intelligence is transforming the way we select embryos in assisted reproduction, improving success rates, reducing risks and providing a less invasive and more accessible way to perform these procedures. While there are still challenges and ethical issues to be addressed, the potential of AI to improve fertility treatments is immense.
The use of this technology could mark a turning point in the way we think about assisted reproduction, offering hope and new possibilities for people seeking to start a family.
References
-Caplan, A. L., et al. (2020). Ethics of artificial intelligence in embryo selection. The American Journal of Bioethics, 20(3), 12-19.
-Morbeck, D. E., et al. (2019). Artificial intelligence in embryo selection: Current applications and future directions. Fertility and Sterility, 112(6), 1057-1066. https://doi.org/10.1016/j.fertnstert.2019.08.015
-Pauli, D., et al. (2020). Time-lapse imaging and AI in embryo selection: Current applications and future perspectives. Reproductive Biomedicine Online, 41(2), 191-199. https://doi.org/10.1016/j.rbmo.2020.04.009
-Rienzi, L., et al. (2017). Preimplantation genetic testing for aneuploidy: A systematic review and meta-analysis of randomized controlled trials. Human Reproduction Update, 23(4), 423-437. https://doi.org/10.1093/humupd/dmx013
-Salumets, A., et al. (2019). The role of artificial intelligence in improving embryo selection in assisted reproduction: A systematic review. Reproductive Biology and Endocrinology, 17(1), 102. https://doi.org/10.1186/s12958-019-0539-x
-Treff, N. R., et al. (2020). Non-invasive genetic testing for embryo selection using cfDNA and AI algorithms. Reproductive BioMedicine Online, 41(3), 527-533. https://doi.org/10.1016/j.rbmo.2020.02.010
-Zhang, W., et al. (2020). Artificial intelligence in embryo selection: Improving the clinical success of in vitro fertilization. Frontiers in Artificial Intelligence, 3, 5. https://doi.org/10.3389/frai.2020.00005
