A couple struggling to have a baby for nearly two decades is finally expecting, thanks to artificial intelligence uncovering sperm previously deemed undetectable.
After 18 years of heartbreak and multiple IVF cycles across various fertility centers, the couple-who wish to remain anonymous-turned to the Columbia University Fertility Center for one last try. The husband was diagnosed with azoospermia, a rare male infertility condition affecting about 10% of infertile men, where no measurable sperm are found in semen.
Dr. Zev Williams, who directs the center, described azoospermia as a “really heartbreaking and unexpected diagnosis,” adding, “Most men feel completely normal, their semen looks normal. But under the microscope, you see no sperm-only fragments and debris.”
The STAR Method
Traditional treatments for azoospermia often involve invasive surgery to extract sperm directly from the testes-a painful and limited option. But a new AI-powered approach called the STAR method is changing this landscape.
STAR, short for Sperm Tracking and Recovery, uses advanced imaging and machine learning to spot hidden sperm cells in semen samples that trained embryologists might never detect.
Dr. Williams explained, “A patient provided a sample, and technicians searched it for two days without finding a single sperm. We used the STAR system, and within an hour, it discovered 44 sperm.”
The technology places the semen on a specialized chip connected to a high-speed camera and microscope. Over an hour, it captures more than 8 million images, rapidly scanning for cells that match the sperm’s unique shape and movement pattern. Once identified, the system isolates each sperm in a micro-droplet, preserving it for IVF.
“It’s like searching for a needle scattered across a thousand haystacks,” Williams said, “but finishing in under an hour, gently enough that the sperm remain usable.”
A milestone pregnancy
In this couple’s case, STAR located three viable sperm, which were used to fertilize eggs through IVF. The pregnancy is progressing, with the baby due this December.
“I still wake up in disbelief,” the wife shared. “After so many failed attempts, I can’t believe it’s finally happening.”
AI in fertility: A transformative force
Artificial intelligence is revolutionising fertility care far beyond sperm detection. Dr. Aimee Eyvazzadeh, a reproductive endocrinologist and host of The Egg Whisperer Show, said, “AI is helping us see what our eyes can’t.”
Several AI tools are already in practice: Stork-A predicts which embryos have the highest chance of success.
CHLOE evaluates egg quality for women considering egg freezing.
AI systems now help tailor IVF medication protocols, optimize sperm selection, and forecast IVF success rates with unprecedented accuracy.
“AI isn’t creating sperm,” Eyvazzadeh emphasised, “but helping us find the rare, viable ones that are already there but nearly invisible. It’s a breakthrough because it amplifies human expertise.”
A more accessible future
The STAR method currently costs around $3,000 and is only available at Columbia University. However, Williams and his team plan to publish their findings to make it accessible to more fertility centers worldwide.
“Infertility is such an ancient struggle-it’s literally biblical,” Williams reflected. “It’s incredible that today’s most advanced technology is being used to solve one of humanity’s oldest problems.”
Looking ahead
Other researchers are also exploring AI in male infertility. In Canada, a team has developed a different AI model to accelerate sperm detection in azoospermic men. Dr. Sevann Helo, a Mayo Clinic urologist unaffiliated with STAR, called these advances “very exciting” and believes AI will “revolutionise the way we look at many medical challenges.”
For countless couples, these breakthroughs mean renewed hope and the possibility of building the families they’ve dreamed about-no matter how hidden the odds.
