A Female-Founded Startup Is Using AI to Improve In Vitro Fertilization Success
Undergoing in vitro fertilization (IVF) is no easy process and therefore it is only untaken by those who deeply desire to have a child. However, these individuals often find their dreams shattered as its success rate ranges between 22 percent and 30 percent, a low number indeed.
Now, a female-founded Israeli startup is claiming that it can increase those numbers through the use of AI. The firm is called Embryonics and its work is nothing short of revolutionary.
To understand what the firm is doing, you must first understand the IVF process. IVF consists of fertilizing a woman’s egg with sperm and creating an embryo in a lab. That embryo is then implanted in the woman's uterus.
After this happens, the woman is forced to take daily hormone shots with unpleasant side effects. Even after all this trouble, very often, the pregnancy does not take place. This is because the embryo chosen by the doctors is not healthy or strong enough.
Although science has advanced on many levels, there is still no better way to check if an embryo is viable other than a doctor's opinion. This is where Embryonics comes in.
The firm is using artificial intelligence (AI) to determine which are the better eggs, ie the ones more likely to lead to a pregnancy. They analyzed 8,789 time-lapse videos of developing embryos and used them to train an algorithm to predict the likelihood of a positive embryo implantation.
And their approach seems to be working. Last week, Tech Crunch reported that in a pilot involving 11 women ranging in age from 20 to 40, six of those had successful pregnancies while the other five were still awaiting results.
That is more than a 50 percent success rate and could end up being even higher if the rest of the women do also end up pregnant. Great news for anyone considering IVF.
Of course, this is not the first time we have witnessed AI being used to grade IVF embryos. Back in 2019, we reported on researchers that trained a Google deep learning algorithm to identify IVF embryos as either good, fair, or poor based on the likelihood each one would successfully implant.
They found that the AI could identify viable embryos as well as an expert could. Embryonics, however, takes this process one step further by using AI to identify successful embryos even better than the experts. Now, that's something to smile about.
Verena Mohaupt, logistics coordinator of MOSAiC, Multidisciplinary drifting Observatory for the Study of Arctic Climate, talks about the perilous journey.