In a groundbreaking development, artificial intelligence (AI) is now being employed to assist fertility doctors in selecting the most viable embryo for in-vitro fertilization (IVF).
With approximately one in every five married women in the United States experiencing difficulties in conceiving after a year of trying, IVF has become a popular solution. According to the Centers for Disease Control and Prevention (CDC), IVF is responsible for 1% to 2% of all births in the country.
However, IVF is not a guaranteed method and can be financially burdensome, with an average cost exceeding $12,000 per session, according to the American Society of Reproductive Medicine.
Also Read
- Over A Quarter Of OECD Jobs At Risk From AI Automation, Workers Fear Job Loss, Says OECD
- “Ethics in Age of Disruptive Technologies” Handbook Guides Ethical Tech Decisions
- Ban On Australia’s School Use Of ChatGPT May Be Reversed, Testing Changes Expected
Furthermore, many women require multiple attempts to achieve pregnancy. Aiming to improve these odds, AIVF, a reproductive technology company based in Tel Aviv, Israel, has developed an AI-powered embryo assessment software called EMA, designed to process vast amounts of data beyond the capabilities of the human eye, streamlining the embryo selection process.
“IVF is one of the most significant medical advancements in the last 50 years, but it’s not good enough,” said Daniella Gilboa, co-founder and CEO of AIVF. “Success rates range from 23% to 25% across all age groups, meaning only one in five IVF attempts leads to pregnancy.”
In the United States, the demand for IVF exceeds the capacity of existing clinics, according to AIVF’s CEO. Gilboa explained that one of the major challenges is that IVF clinics cannot keep up with the growing demand.
“More women are choosing to freeze their eggs to delay childbirth and focus on their careers, resulting in an increasing demand for IVF. However, the supply is limited,” she said. “In the U.S., only 20% of the demand is being met, which means 80% of these women are giving up on the dream of having a child.”
Embryo selection is one of the most critical decision points in the IVF process. Traditionally, human clinicians have relied on their personal judgment to make this choice.
Gilboa described the challenging scenario for embryologists, who must evaluate multiple embryos in a busy laboratory environment and determine which one has the highest potential for successful development. With AIVF’s EMA software, an AI model assists clinicians by evaluating embryos and applying advanced algorithms to identify the most promising candidates for a successful pregnancy.
“The AI has been trained to detect embryonic features that correlate with different outcomes, such as genetic abnormalities, implantation potential, or gender, which may not be visible to the human eye,” Gilboa explained. Each embryo is assigned a numeric score, and the clinician can then make the final selection.
Without AI, clinicians would have to rely solely on the appearance of the embryo, which is subjective and does not accurately quantify the chances of pregnancy. “The AI allows the doctor to reduce uncertainties and provide clear, accurate information,” she added.
Furthermore, the AI system operates much faster than humans, evaluating embryos in a fraction of the time. This increased efficiency means that clinics can see more patients and meet the rising demand for IVF services.
EMA’s AI software was trained using a large dataset of time-lapse videos capturing embryos’ development, analyzing which ones resulted in successful pregnancies and which ones did not.
“It was a lengthy process with multiple stages to construct a model that was accurate enough to begin using,” Gilboa stated, emphasizing the importance of extensive data collection.
Experts in the field of reproductive medicine have welcomed AI technologies like AIVF’s as valuable tools to overcome the limitations of manpower and expertise in the fertility industry.
Dr. Shahin Ghadir, a double board-certified infertility specialist and Trimly fertility coach in California, expressed optimism about the use of AI in embryology. “I think embryologists are still needed in the laboratory, but combining their knowledge with the use of artificial intelligence can be highly beneficial,” he said.
However, Gilboa emphasized that EMA is not intended to replace doctors but to assist them. “At the end of the day, the human makes the decision, not the AI,” she affirmed.
“It’s only a tool that enables clinicians to evaluate the embryos, consult with the patients, and provide better IVF care.” She concluded, “It’s not ‘humans versus AI’—it’s ‘humans with AI’.”
