Flawed or non-existent sperm tests mean that the burden of fertility treatments too often falls on women –Mojo is looking to change that using an AI-powered microscope
Mohamed Taha’s ambition to disrupt the fertility industry began after his first sperm test. He had just been diagnosed with a kidney disease (which later turned out to be a misdiagnosis) and his doctor advised him to freeze his sperm as a precaution. According to the World Health Organisation, a normal sperm count is around 15 million sperm per millilitre (m/ml) of semen. Taha’s sperm count was 15 times lower.
Concerned, he chose to do a second analysis at a different clinic and, to his surprise, the result was far more positive: 20 m/ml. A third analysis, however, gave him a sperm count of 3 m/ml. “It was frustrating,” he says. “Every time I did a sperm test I would get a different result.”
Taha started asking around and eventually found out the reasons for the discrepancy: most fertility clinics don’t employ doctors who specialise in male reproductive systems (known as andrologists); sperm analyses often don’t follow the gold standard WHO protocol; these tests are also often done manually, using antiquated equipment.
“After speaking to doctors, they told me that, as a man, I shouldn’t really worry about my fertility – in the worst-case scenario all that they needed from me was one sperm, and that my partner could get an in-vitro fertilisation (IVF) treatment if necessary.” Taha found this unacceptable. “We’ve all heard the horror stories about IVF,” he says.
During IVF, women are injected with potent hormonal treatments to stimulate egg production. These, however, can have serious side effects, ranging from depression to cardiac complications and even death. “Couples are often poorly informed,” he says.
According to Taha, few are told that the overall chance of success for IVF is only 25 per cent. Fewer know that in half those cases IVF doesn’t work because there’s a problem with the man’s sperm. To make matters worse, these problems are often only uncovered after several rounds of unsuccessful IVF treatments: in 90 per cent of the cases, men are never assessed properly by an andrologist when they first attend a fertility clinic.
Taha, who was doing a PhD in nanotechnology at the time, thought he could do better by applying artificial intelligence and nanorobotics to the problem of human fertility. He founded Mojo in 2017, with co-founders Daniel Thomas, Fanny Chesa and Tobias Boecker. “We wanted to get rid of the human error factor entirely,” he says. “We want to replace an entire IVF lab with intelligent robots that could do it faster, cheaper and more efficiently.”
By 2018, Mojo had developed its first product: a smart microscope that could automatically scan a sample of sperm and analyse it. Called Mojo Assess, it is powered by a computer vision algorithm trained on millions of sperm images labeled by expert andrologists. It can compute sperm count, sperm motility, the predominant shape of the sperm and other parameters like the integrity of the sperm’s DNA. It has been tested in clinical trials conducted at the best fertility institutes in the world, including The Doctors Laboratory, King’s Fertility, and the Karolinska Institute, in Solna, Sweden.
According to these tests, Mojo Assess results show a 97 per cent agreement with the gold standard analysis done according to the WHO guidelines. Furthermore, Mojo’s technology can perform a test in four minutes, whereas WHO protocols, on average, typically take 30 minutes and two lab technicians to complete. “Healthy sperm is half of a successful IVF process,” Taha says. “When properly evaluated and selected, we increase its chance of success and prevent women from undertaking unnecessary treatments.”
To Taha, Mojo Assess is just the first step. The company is currently developing a robot, called Mojo Inject, that will automate sperm injection. This robot will use AI to select healthy sperm and inject it into an egg with precise micrometric robotic needles and advanced microfluidics. “It’s a tricky procedure that can only be done by a technician with decades of experience, and it requires much precision not to destroy the eggs, which are scarce,“ Taha says. “Sadly, there’s no gold standard protocol for this.”
Taha estimates that, when used together, Mojo’s suite of products can increase the probability of IVF success from the current 25 per cent to 50. “Unfortunately, it’s biologically impossible for some couples to have a child,” he says. “Our promise is that, using our technology, if a couple can have a baby, they will probably have it.”