AI Tools Transform Embryo Selection in IVF Procedures

Artificial intelligence tools are increasingly being used to enhance the selection of embryos in in vitro fertilisation (IVF) procedures. These advancements promise to improve success rates for couples struggling to conceive, although they also raise ethical considerations that warrant attention.

In IVF, embryos are created by fertilising an egg with sperm in a laboratory. Despite the procedure allowing millions of families to welcome children each year, its success rates can vary significantly and tend to decline with maternal age. Nearly five decades after the birth of the first IVF baby, experts like Nathalie Massin, the head of the clinical unit at the American Hospital of Paris, emphasize that “artificial intelligence is here to help us select better embryos or at least help determine their potential for implantation.” The hospital performs over 2,300 IVF cycles annually, utilising advanced technologies to improve outcomes.

AI Enhances Embryo Assessment

At the American Hospital of Paris, an embryoscope serves as a time-lapse camera, continuously monitoring embryo development. Previously, the data collected—such as the shape, symmetry, and cell division patterns—was underutilised. Now, AI-driven tools, particularly those employing machine learning, assist doctors in identifying embryos with the highest likelihood of successful implantation or freezing for future use. This innovation aims to reduce the number of costly IVF attempts by eliminating embryos most likely to exhibit genetic abnormalities that can lead to miscarriages.

Frida Entezami, co-leader of the IVF department at the same hospital, noted that while AI plays a crucial role, “humans will continue to make the decisions but with this additional tool.” The AI system, developed by Israeli start-up AIVF, is currently undergoing internal testing, with the goal of halving the number of cycles necessary to achieve pregnancy. Entezami highlighted that AIVF’s technology boasts a 70 percent probability of identifying embryos free from genetic abnormalities, a significant improvement considering that approximately half of pre-implantation embryos exhibit such issues.

Personalized Support and Ethical Considerations

Additionally, AI tools can optimize hormone injection protocols prior to egg retrieval and enhance the chances of identifying sperm in samples with low counts. According to Anne-Claire Lepretre, head of the assisted reproductive technologies unit at France’s Biomedicine Agency, these algorithms are rigorously tested to ensure accuracy. When IVF attempts do not succeed, data from these experiences can be integrated into AI models, increasing the potential for future success.

Lepretre believes that such personalised assistance could help “avoid the emotional rollercoaster” experienced by many couples on their journey to parenthood. She described the IVF process as “long, complex, and often psychologically difficult.”

Despite these advancements, some researchers urge caution regarding the ethical implications of using AI in embryo selection. Julian Koplin, a bioethicist at Monash University, expressed concerns that AI is beginning to dictate decisions on who may be born. In a statement, he mentioned that individuals with moral objections to AI-assisted decisions should be made aware of its application and ideally have the option to opt out. His review, published in the journal Human Reproduction, called for careful scrutiny but concluded that ethical concerns do not necessarily oppose the use of AI in embryo assessment.

As the field of reproductive technology evolves, the integration of AI into IVF procedures represents a significant shift. Experts emphasize the need for ongoing discussions about the implications of such technologies, ensuring that advancements remain aligned with the values and preferences of prospective parents.