After 18 years of trying to conceive without success due to infertility, a remarkable breakthrough has occurred: artificial intelligence (AI) has identified 44 viable sperm cells in a man who had been previously deemed infertile, and his wife is now expecting a child.
Reports indicate that the AI was able to locate the viable sperm within just one hour of analysis, despite the man being classified as infertile for nearly two decades.
Researchers at Columbia University Fertility Centre in the United States have revealed that they utilised AI to detect viable sperm in a man diagnosed with infertility.
The Washington Post reported on Friday that the man, who suffered from azoospermia—a condition marked by an extremely low or absent sperm count—and his wife had been attempting to conceive for 18 years.
According to TODAY, the woman underwent the first successful embryo transfer in March 2025, using cells retrieved through this groundbreaking AI process, and she is now pregnant.
How the AI Sperm Detection Works
Zev Williams, the lead researcher and director of the university’s fertility programme, explained that laboratory technicians had analysed the patient’s semen sample for two days without detecting any sperm. However, after employing an AI-powered system, they were able to identify 44 viable sperm cells in under an hour.
“Labs searched for two days and found nothing. We ran the same sample and found 44 sperm within an hour,” he stated. “That changes everything for a couple who thought they had no path forward.”
The system, dubbed Sperm Track and Recovery (STAR), integrates an AI algorithm for detecting sperm with a fluidic chip that passes the semen sample through a tiny tubule on a plastic chip. As the sample progresses, a microscope connected to a high-speed camera captures millions of images in quick succession.
These images are then scanned in real time by AI software trained to detect viable sperm cells. Once identified, the sperm can be preserved through freezing or utilised in in-vitro fertilisation (IVF) via intracytoplasmic sperm injection (ICSI), a method where a single sperm is injected directly into an egg.
Researchers indicated that this method reduces the time spent searching for sperm and helps avoid invasive surgical procedures typically required to extract sperm directly from the testicles.
Expert Opinions
Robert Brannigan, the president-elect of the American Society for Reproductive Medicine, has described this innovation as “promising.”
“Even the discovery of a single viable sperm can be life-changing,” he remarked.
Allison Rodgers, a reproductive endocrinologist at Fertility Centres of Illinois, noted that these findings reveal significant gaps in current fertility methods.
She believes that AI will be pivotal in the future of IVF. “It’s amazing and makes me realise that perhaps what we thought was advanced still has a long way to go,” she stated. “I believe AI is going to absolutely revolutionise IVF.”
However, Gianpiero Palermo, an infertility specialist at Weill Cornell Medicine, has called for caution and stressed the necessity for further validation of the technology.
“You are attracting patients who have been told they have no sperm and offering what may turn out to be false hope,” Palermo warned.
Williams and his team are currently exploring ways to adapt the system for additional applications, including the identification of healthy sperm, eggs, and embryos.