Skip to content 
After nearly two decades of unsuccessful attempts to conceive, including 15 failed rounds of in vitro fertilisation (IVF), a U.S.-based couple has finally achieved pregnancy, marking the first documented success of an innovative artificial intelligence system developed at Columbia University. The new procedure, known as the STAR method, uses AI to detect rare, viable sperm in men diagnosed with a severe form of infertility known as azoospermia.
AI Breakthrough in Fertility Care
The STAR method, short for Sperm Tracking and Recovery, was developed over five years by Dr. Zev Williams and his team at the Columbia University Fertility Centre. The technology combines a fluidic chip with an AI algorithm that scans semen samples for live sperm in cases where none are visible through conventional methods. The AI-assisted system can analyze a semen sample under a microscope using high-speed imaging, taking more than 8 million frames in less than an hour. It then isolates any viable sperm into droplets, preserving them for use in IVF.
According to Williams, the AI system significantly outperforms manual examination. In one instance, embryologists spent two days reviewing a sample and found no sperm. However, when the same sample was passed through the STAR system, it detected 44 sperm within an hour. This demonstrated a considerable leap in accuracy and efficiency, offering a new option to patients previously left with few or no alternatives.
First Successful Pregnancy Using STAR
The couple, who chose to remain anonymous, had been trying to conceive for 18 years. The male partner was diagnosed with azoospermia, a condition affecting around 10% of infertile men, in which semen contains little to no sperm. This condition often goes unnoticed, as semen samples appear normal in volume and texture but contain no detectable sperm under a microscope.
Traditional treatment options for azoospermia typically involve painful and invasive surgical procedures to extract sperm directly from testicular tissue. Hormonal therapies may help only when hormone imbalances are the underlying cause. In cases where no sperm are found, couples are left with the option of using donor sperm.
After exhausting these avenues, the couple turned to the Columbia University Fertility Centre. Using the STAR system, technicians found three viable sperm in the male partner’s sample. These were then used to fertilize the female partner’s eggs through IVF. The resulting embryo was successfully implanted, and the couple is now expecting their first child, due in December.
“It took me two days to believe I was pregnant,” the woman said in an interview with CNN. “I still wake up in the morning and can’t believe if this is true or not.”
Also read: U.S. Leads the Global AI Race
Artificial Intelligence Expands Its Role in Fertility Medicine
Artificial intelligence has been gradually integrated into various stages of fertility care. Existing applications include embryo screening to determine the healthiest embryos for transfer and the evaluation of egg quality before freezing. Tools like Stork-A and CHLOE are already being used in IVF clinics to enhance embryo and egg assessment.
Dr. Aimee Eyvazzadeh, a reproductive endocrinologist based in San Francisco, noted that AI enables clinicians to make better decisions and improve patient outcomes. She explained that systems like STAR amplify human expertise rather than replace it, allowing embryologists to recover rare sperm that may have otherwise gone unnoticed.
“AI isn’t creating sperm—it’s helping us find the rare, viable ones that are already there but nearly invisible,” Eyvazzadeh said in an email.
The STAR method, specifically tailored to address male infertility, now brings this technological advancement into a previously under-addressed area of reproductive medicine.
Validation and Cost Considerations
Despite its groundbreaking success, some experts urge caution. Nevertheless, some researchers are calling for temperance. Dr. Gianpiero Palermo is a professor of embryology at Weill Cornell Medicine who stressed that AI in reproductive medicine has yet to be thoroughly validated. Palermo, the pioneer of the intracytoplasmic sperm injection (ICSI) technique, which is currently widely employed in IVF, noted that human embryologists still need to choose and inject the sperm into the egg. He admitted that despite the possible assistance that AI could provide in identifying more sperm in a shorter period, AI would not be able to alter the root problem in the scenario when there are no sperm available.
The procedure can aid in the retrieval of sperm slightly quicker and possibly get an extra one than the embryologist, Palermo added. However, populations of spermatozoa will be absent in some men, and there is nothing that the screening procedure can do at all.
Dr. Williams accepted the fact that the STAR system can be found exclusively at the Columbia University Fertility Centre at the moment. He (and his team) will, however, publish their studies and share the technology with fertility centres in different parts of the world. The estimated cost of using STAR system in sperm identification, isolation and freezing is less than 3000 dollars.
Artificial Intelligence Emerging Treatment of Male Infertility
Artificial intelligence in the treatment of male infertility may become a milestone in changing diagnostic and treatment methodology of reproductive problems. The Centres for Disease Control and Prevention (CDC) indicate that male factors are among the leading causes of infertility among couples, affecting about 40 per cent of this family group. Nevertheless, greater attention in the discussions and medical management has been given to the problems of female reproduction.
Azoospermia is a very hard condition to deal with since it is not externally evident. The majority of men affected by azoospermia, as Williams explained, are healthy; they can have normal sex, and their semen has a normal appearance. The only indicator of infertility is with the help of a microscope where no sperm is observed.
The thing is that, unlike the standard number of sperm (200 million to 300 million in an average sample), these patients are likely to have only two or three, wrote Williams. But the specificity of STAR system combined with the experience of our embryologists, even such a small number of people can be used to fertilize an egg successfully.
The Future Prospects
The effectiveness of the STAR technique is a tremendous breakthrough in fertility therapy and speaks volumes of AI development in the healthcare sector. With the researchers now poised to offer the technology to more people, they are still keen to make sure that the technology passes the test of proper scientific validation and ethical usage.
Williams has hope that STAR will spread and be easily accessible to other couples that are unable to have children due to otherwise seemingly unsurpassable fertility difficulties.
Infertility is a kind of unique in the sense that it is one of the oldest things in human existence. It is biblical, he said. It is incredible that the most sophisticated technologies which we at the present have are set to cure this old issue.
The story of the couple is a testimony of how science and patience can work together to bring new life on the verge of the birth deadline.
Comments 1