Researchers achieve first pregnancy using AI to recover sperm in infertile men

Researchers have reported the first successful pregnancy using an AI-guided sperm recovery method in infertile men with azoospermia, a condition where ejaculate contains little or no sperm.

The case involved a man who had spent nearly 20 years trying to start a family, undergoing multiple IVF cycles, manual sperm searches and two surgical procedures.

Male factors account for around 40 per cent of infertility in couples, and 10–15 per cent of infertile men have azoospermia.

Researchers at Columbia University Fertility Center developed the technique, detailed in a research letter published online on 31 October 2025.

Zev Williams is senior author and director of the Columbia University Fertility Center.

Williams said: “A semen sample can appear totally normal, but when you look under the microscope you discover just a sea of cellular debris, with no sperm visible.”

Many couples with male-factor infertility are told they have little chance of having a biological child.

Men with azoospermia may undergo a procedure to have sperm surgically extracted from the testes, but the procedure is often unsuccessful and can cause vascular problems, inflammation or a temporary drop in testosterone.

Currently, a few specialist labs manually inspect semen samples — a slow and expensive process — often after centrifugation or other treatments that can damage sperm.

“The field has really been challenged to find a better way to identify and retrieve viable sperm cells in men with exceedingly low sperm counts,” Williams said.

The new approach, called STAR (Sperm Tracking and Recovery), combines imaging, microfluidics and AI to identify and recover rare sperm cells from men with azoospermia.

The method uses high-powered imaging to scan semen samples, taking more than eight million images in under an hour.

AI then pinpoints sperm cells, and a microfluidic chip – a device with tiny, hair-like channels – isolates the portion containing sperm.

Within milliseconds, a robot gently removes the sperm so it can be used to create an embryo or frozen for later use.

Hemant Suryawanshi is assistant professor of reproductive sciences at Columbia University Vagelos College of Physicians and Surgeons and project leader.

Suryawanshi said: “Our team included experts in advanced imaging techniques, microfluidics, and reproductive endocrinology to tackle each step required to find and isolate rare sperm.”

STAR was tested in a patient who provided a 3.5 mL semen sample.

Within about two hours, it scanned 2.5 million images and identified two viable sperm cells, which were used to create two embryos and start a pregnancy.

“You only need one healthy sperm to create an embryo,” Williams said.

Although based on a single case, the findings demonstrate the feasibility of using this technology to help men with azoospermia overcome long-standing barriers to biological parenthood.

Larger clinical studies are now under way to assess STAR’s efficacy in broader patient populations.

The research team first unveiled the STAR method earlier this year, combining advanced imaging and microfluidic technologies to detect and extract rare sperm cells.

Once identified, the system can remove sperm within milliseconds for immediate use or freezing.