How AI Enhances IVF Success Rates: The Latest Research

New artificial intelligence technology can improve IVF success rates by determining which embryos have the best chance of implantation.

Artificial intelligence (AI) can assist healthcare professionals by analyzing MRI or CT scans, assessing potential treatment options, and interpreting medical record data.

For example, AI models can detect tumor margins better than an MRI and help diagnose and develop treatment plans for prostate cancer.

Now, a new AI tool developed by AIVF — an Israeli company founded by embryologist Daniella Gilboa and IVF expert Professor Daniel Seidman — can potentially increase in vitro fertilization (IVF) success rates by as much as 30%.

In a study published in Advanced Science, Gilboa, Seidman, and other scientists examined whether their new AI technology — called EMA — could analyze information encoded within "sibling" embryos from the same IVF group and predict which had the best chance of successfully implanting in the uterus.

Daniella Gilboa told Healthnews, "The main dogma in IVF is that each embryo in the cohort is independent, meaning that one embryo does not affect the chances of success of the other embryo. We showed that this is incorrect, as a cohort of bad embryos has a lower chance of implantation, even if this cohort constitutes one good embryo."

Gilboa says that the chance of successful implantation of this specific good embryo is lower than the same embryo coming from a more viable group. This means embryos in the same group have some dependency on each other.

"The research also proved this as AIVF has been developing these models for four years, based on a huge amount of IVF data," Gilboa said. "AIVF's models recognize features on the developing embryo that are correlated with different outcomes like successful pregnancy and genetic abnormalities, and are able to assist in predicting which embryo has the highest chances of succeeding, to reduce the uncertainty of the treatment and reduce time to pregnancy."

According to Gilboa, the AIVF system (EMA) works as the operating system of an IVF clinic. It collects embryo and patient data and analyzes it through different AI models, assisting embryologists and physicians with making decisions throughout the treatment. The system also automates clinic processes such as reporting, documenting, and quality assurance and control. In addition, it provides remote 24/7 analysis of the embryos, Key Performance Indicator (KPI) monitoring, and communication.

Currently, AIVF operates in fertility clinics in Europe, Australia, India, and Brazil. Gilboa says the system will be available in the United States in a few weeks.

Other companies that have developed AI-powered IVF technologies include ImVitro, Alife, Fairtility, and Embryonics.

What is IVF?

In vitro fertilization, commonly known as IVF, is an assisted reproductive technology (ART) designed to help couples experiencing infertility conceive a child.

IVF involves the fusion of an egg and sperm outside the human body within a laboratory setting, creating an embryo. This embryo is then carefully monitored and implanted into the uterus, where it has the potential to develop into a healthy pregnancy.

Medical professionals begin IVF by stimulating the ovaries to produce more eggs. Then, the eggs are combined with sperm in a controlled laboratory setting. To facilitate fertilization, lab technicians sometimes inject sperm directly into an egg.

The fertilized eggs are allowed to develop into embryos and then placed into the uterus 3-5 days after fertilization.

However, this process can be time-consuming, costly, and emotionally and physically draining for couples. Therefore, improving IVF success rates using AI could be a less stressful and more effective solution for people with fertility challenges.