A couple tried for 18 years to get pregnant. AI made it happen

CNN

03-07-2025

AI
Maternal health
Women's health
FacebookTweetLink
After trying to conceive for 18 years, one couple is now pregnant with their first child thanks to the power of artificial intelligence.
The couple had undergone several rounds of in vitro fertilization, or IVF, visiting fertility centers around the world in the hopes of having a baby.
The IVF process involves removing a woman's egg and combining it with sperm in a laboratory to create an embryo, which is then implanted in the womb.
But for this couple, the IVF attempts were unsuccessful due to azoospermia, a rare condition in which no measurable sperm are present in the male partner's semen, which can lead to male infertility. A typical semen sample contains hundreds of millions of sperm, but men with azoospermia have such low counts that no sperm cells can be found, even after hours of meticulous searching under a microscope.
So the couple, who wish to remain anonymous to protect their privacy, went to the Columbia University Fertility Center to try a novel approach.
It's called the STAR method, and it uses AI to help identify and recover hidden sperm in men who once thought they had no sperm at all. All the husband had to do was leave a semen sample with the medical team.
'We kept our hopes to a minimum after so many disappointments,' the wife said in an emailed statement.
Researchers at the fertility center analyzed the semen sample with the AI system. Three hidden sperm were found, recovered and used to fertilize the wife's eggs via IVF, and she became the first successful pregnancy enabled by the STAR method.
The baby is due in December.
'It took me two days to believe I was actually pregnant,' she said. 'I still wake up in the morning and can't believe if this is true or not. I still don't believe I am pregnant until I see the scans.'
Artificial intelligence has advanced the field of fertility care in the United States: More medical facilities are using AI to help assess egg quality or screen for healthy embryos when patients are undergoing IVF. There's still more research and testing needed, but AI may now be making advancements in male infertility, in particular.
Dr. Zev Williams, director of the Columbia University Fertility Center, and his colleagues spent five years developing the STAR method to help detect and recover sperm in semen samples from people who had azoospermia.
They were struck by the system's results.
'A patient provided a sample, and highly skilled technicians looked for two days through that sample to try to find sperm. They didn't find any. We brought it to the AI-based STAR System. In one hour, it found 44 sperm. So right then, we realized, 'Wow, this is really a game-changer. This is going to make such a big difference for patients,' ' said Williams, who led the research team.
When a semen sample is placed on a specially designed chip under a microscope, the STAR system – which stands for Sperm Tracking and Recovery – connects to the microscope through a high-speed camera and high-powered imaging technology to scan the sample, taking more than 8 million images in under an hour to find what it has been trained to identify as a sperm cell.
The system instantly isolates that sperm cell into a tiny droplet of media, allowing embryologists to recover cells that they may never have been able to find or identify with their own eyes.
'It's like searching for a needle scattered across a thousand haystacks, completing the search in under an hour and doing it so gently, without any harmful lasers or stains, that the sperm can still be used to fertilize an egg,' Williams said.
'What's remarkable is that instead of the usual [200 million] to 300 million sperm in a typical sample, these patients may have just two or three. Not 2 [million] or 3 million, literally two or three,' he said. 'But with the precision of the STAR system and the expertise of our embryologists, even those few can be used to successfully fertilize an egg.'
It's estimated that the male partner accounts for up to 40% of all infertility cases in the United States, and up to 10% of men with infertility are azoospermic.
'This often is a really heartbreaking and shocking and unexpected diagnosis,' Williams said. 'Most men who have azoospermia feel completely healthy and normal. There's no impairment of their sexual function, and the semen looks normal, too. The difference is that when you look at it under a microscope, instead of seeing literally hundreds of millions of sperm swimming, you just see cell debris and fragments but no sperm.'
Treatment options for azoospermia traditionally have included uncomfortable surgery to retrieve sperm directly from a patient's testes.
'A part of the testes gets removed and broken into little pieces, and you try to find sperm there,' Williams said. 'It's invasive. You can only do it a couple of times before there could be permanent scarring and damage to the testes, and it's painful.'
Other treatment options may include prescription hormone medications – but that will be effective only if the person has an imbalance of hormones. If no other treatment options are successful, couples may use donor sperm to have a child.
Williams said the STAR method can be a new option.
'It really was a team effort to develop this, and that's what really drove and motivated everybody, the fact that you can now help couples who otherwise couldn't have that opportunity,' he said.
Although the method is currently available only at the Columbia University Fertility Center, Williams and his colleagues want to publish their work and share it with other fertility centers. Using the STAR method to find, isolate and freeze sperm for a patient would cost a little under $3,000 total, he said.
'Infertility is unique in a way in that it's such an ancient part of the human experience. It's literally biblical. It's something we've had to contend with through all of human history,' he said. 'It's amazing to think that the most advanced technologies that we currently have are being used to solve this really ancient problem.'
It's not the first time doctors have turned to AI to help men with azoospermia.
A separate research team in Canada built an AI model that could automate and accelerate the process of searching for rare sperm in samples from men with the condition.
'The reason AI is so well-suited for this is AI really relies on learning – showing it an image of what a sperm looks like, what the shape is, what characteristics it should have – and then being able to use that learning algorithm to help identify that specific image that you're looking for,' said Dr. Sevann Helo, a urologist at Mayo Clinic with specialty interest in male infertility and male sexual dysfunction, who was not involved in the STAR method or the research in Canada.
'It's very exciting,' she said. 'AI, in general, at least in the medical community, I think is a whole new landscape and really will revolutionize the way we look at a lot of problems in medicine.'
The STAR method is a novel approach to identifying sperm, but AI has been used in many other ways within fertility medicine too, said Dr. Aimee Eyvazzadeh, a San Francisco-based reproductive endocrinologist and host of the podcast 'The Egg Whisperer Show.'
'AI is helping us see what our eyes can't,' Eyvazzadeh, who was not involved in the development of STAR, wrote in an email.
For instance, AI algorithms, such as one called Stork-A, have been used to analyze early-stage embryos and predict with 'surprising accuracy' which ones are likely to be healthy. Another AI tool, CHLOE, can assess the quality of a woman's eggs before she may freeze them for future use.
'AI is being used to personalize IVF medication protocols, making cycles more efficient and less of a guessing game. It's also helping with sperm selection, identifying the healthiest sperm even in difficult samples. And AI can now even predict IVF success rates with more precision than ever before, using massive data sets to give patients personalized guidance,' Eyvazzadeh said. 'The common thread? Better decisions, more confidence, and a more compassionate experience for patients.'
The new STAR system is 'a game-changer,' she said.
'AI isn't creating sperm – it's helping us find the rare, viable ones that are already there but nearly invisible,' she said. 'It's a breakthrough not because it replaces human expertise, but because it amplifies it – and that's the future of fertility care.'
But there is also a growing concern that the rushed application of AI in reproductive medicine could give false hope to patients, said Dr. Gianpiero Palermo, professor of embryology and director of andrology and assisted fertilization at Weill Cornell Medicine.
'AI is gaining a lot of traction nowadays to offer unbiased evaluation on embryos by looking at embryo morphology,' Palermo said in an email. 'However, current available models are still somewhat inconsistent and require additional validation.'
Palermo said the STAR approach needs to be validated and would still require human embryologists to pick up sperm and inject them into an egg to create an embryo for patients undergoing IVF.
'Maybe the AI addition may help to retrieve the spermatozoon a little faster and maybe one more than the embryologist,' said Palermo, who was not involved in the development of STAR but was the first to describe the method of injecting sperm directly into an egg. Since he pioneered that method, it has become the most-utilized assisted reproductive technology in the world.
'In my opinion, this approach is faulty because inevitably some men will have no spermatozoa,' Palermo said of the STAR method, 'doesn't matter how their specimens are screened whether by humans or a machine.'