After 19 years of trying for a baby, this couple lost hope. But AI made it happen

Artificial intelligence is making swift strides in healthcare in a short time — and now, in a first, it has even helped a couple get pregnant. After 19 years of trying to start a family, with azoospermia (no detectable sperm in ejaculate) and 15 failed IVF cycles, the couple finally became pregnant in March this year after using what is called a Sperm Track and Recovery (STAR) system.
Just like astrophysicists use AI to find new stars, Columbia University Fertility Center researchers developed STAR to detect “really, really, really rare sperm”, the kind lab technicians fail to find in azoospermia samples. “I liken it to finding a needle hidden within a thousand haystacks. But it can do that in a couple of hours — and so gently that the sperm that we recover can be used to fertilise an egg,” Dr Zev Williams, STAR developer and fertility centre director told Time Magazine.

For five years, Williams and his team perfected the system using a sperm-detecting AI algorithm. The process starts with a fluidic chip flowing the semen sample through a small tube on a plastic chip. Once AI confirms sperm presence, that bit of semen is diverted to a separate tube. Whatever few sperm there are in the sample can be isolated in such a way, and either used to fertilise an egg, or frozen for later use of the couple.

That’s exactly what Williams and his team did for the couple. Just two hours after collecting the husband’s sperm, they got to know the wife’s eggs had been successfully fertilised, ready for transfer to the uterus in a few days. The wife is four months pregnant now and both the mother and fetus are doing well.

Columbia University Fertility Center says the STAR system is a groundbreaking advancement for men diagnosed with azoospermia, adding, “This system can detect and retrieve even the smallest numbers of sperm gently and without harsh chemicals or lasers, using cutting-edge AI, high-speed imaging, and robotics.”

What makes STAR ahead of other AI detecting systems is that it can also successfully isolate the sperm in a semen sample where male infertility is involved. “As a test, we decided to run samples where embryologists could not find any sperm through the system. The embryologists really worked hard to find sperm, since they didn’t want to be outshone by a machine. In one of the samples, they analysed for two days and found no sperm. STAR found 44 in an hour,” says Williams.

Azoospermia affects around 1% of all men and accounts for 10% of all male fertility issues. The most common cause of the condition is a blockage in the male reproductive tract. Other causes include genetic conditions, hormonal imbalances, endocrine disorders, radiation therapy, environmental toxins, and drug or alcohol abuse. Williams says azoospermia is one of the many fertility issues that his AI STAR system could find a solution to. “The dream is to develop technologies so that those who are told ‘you have no chance of being able to have a child’ can now go on to have healthy children,” he adds. Agencies