Scientists at the Columbia University Fertility Center (United States) have reported the first successful pregnancy achieved using an Artificial Intelligence (AI)-guided method designed to retrieve sperm from men with azoospermia, in whom the ejaculate contains little or no sperm.

The case is described in A research letter published this Saturday (November 1, 2025) in The Lancet.

Research letters report research results and are externally peer-reviewed, but unlike articles, they are shorter and usually report preliminary studies or initial findings.

Male factors account for about 40% of infertility cases in couples, and of these, between 10 and 15% are men with azoospermia.

A semen sample may appear completely normal, but when viewed under a microscope, “it reveals a sea of ??cellular debris, with no sperm visible,” explains Zev Williams, lead author of the article and director of the Columbia University Fertility Center.

Therefore, many couples with male factor infertility have little chance of having a biological child.

Men with azoospermia can undergo a procedure to surgically extract sperm from the testicles, but the operation is often unsuccessful and can cause vascular problems, inflammation, or a temporary decrease in testosterone levels.

Some specialized laboratories employ technicians to manually inspect semen samples, but the lengthy and expensive process can damage sperm.

“The field has faced the challenge of finding a better way to identify and recover viable sperm in men with extremely low sperm counts,” Williams explains.

A healthy sperm

To find an alternative, Williams assembled a multidisciplinary team of researchers and physicians to devise a new method combining various technologies to recover sperm from men with azoospermia. The STAR (Sperm Tracking and Recovery) method, unveiled earlier this year, uses high-powered imaging technology to scan a semen sample from men with azoospermia (taking more than 8 million images in under an hour). The sperm in the sample are then identified using AI,and a microfluidic chip with tiny, hair-like channels isolates the portion of the semen sample containing the sperm. In milliseconds, a robot extracts the sperm so they can be used to create an embryo or frozen and stored for future use. STAR was tested on a patient who had been trying to start a family for nearly 20 years and had undergone Multiple in vitro fertilization cycles at other centers, several manual sperm searches, and two surgical procedures to extract sperm.

The patient provided a 3.5 ml semen sample. In about two hours, STAR scanned 2.5 million images and identified two viable sperm, which were then used to create two embryos and achieve a pregnancy.

The results, although based on a single case, show the viability of this technology to overcome the barriers that have long prevented helping men with azoospermia.

“It only takes one healthy sperm to create an embryo,” says Williams.

Larger clinical studies are currently underway to evaluate the effectiveness of STAR in broader patient populations.