- By Catherine Latham
- Technology Journalist
Infertility affects 7% of the male population. Now Artificial Intelligence (AI) is going to help solve the problem.
Dr Steven Vasilescu says the AI software he and his team developed can detect sperm in samples taken from severely infertile men 1,000 times faster than a pair of highly trained eyes.
“It can highlight potentially viable sperm before the man processes what he’s seeing,” he says.
Dr Vasilescu is a biomedical engineer at the University of Technology Sydney (UTS) in Australia and the founder of the medical company Neogenics Biosciences.
The system he and his colleagues developed is called SpermSearch.
It is designed to help men who have no sperm in their ejaculate, the 10% of infertile men, with a condition called non-obstructive azoospermia (NOA).
Usually in these cases, a small part of the testicle is surgically removed and taken to the laboratory, where the embryologist can look for healthy sperm themselves.
The tissue is separated and examined under a microscope. If any viable sperm are found, they can be collected and injected into the egg.
The process can take many workers six or seven hours, Dr Vasilescu says, and is prone to fatigue and error.
“When an embryologist looks under the microscope, what they see is just this whole mess—a starscape of cells,” he says.
“There’s blood and tissue. There may be only 10 sperm in the whole thing, but there may be millions of other cells. It’s a needle in a haystack,” says Dr Vasilescu.
He says that, by contrast, SpermSearch can detect any healthy sperm in seconds, when pictures of the samples are instantly uploaded to a computer.
To achieve this speed, Dr Vasilescu and his colleagues trained AI to identify sperm in these complex tissue samples by showing them thousands of such images.
In a published scientific paper, the UTS biomedical engineering team said that in one test SpermSearch was 1,000 times faster than an experienced embryologist.
However, SpermSearch is not designed to replace embryologists, but to act as an auxiliary tool.
Dr. Sarah Martins da Silva says that speed is key to finding any sperm. “Time is critical,” says a clinical reader in reproductive medicine at the University of Dundee.
“If you have someone else store eggs and you have eggs that need to be fertilized, there’s only a short amount of time for us to be able to do that. Speeding up the process would be very beneficial.”
Infertility is a growing problem, with sperm counts reported to have halved over the past four decades.
Dr Meurig Gallagher does another educational work to help men with infertility problems.
An assistant professor in the Center for Systems Modeling and Quantitative Biomedicine at the University of Birmingham, his new technique uses imaging software to track the speed and movement of sperm tails.
“Looking at the tail gives information about the health of the specimen,” he says. “Minute changes can tell us whether a sperm is under environmental stress, dying or responding to a biological signal.”
Meanwhile, Belfast-based fertility firm Examine uses a technique known as single cell gel electrophoresis to detect DNA damage in individual sperm.
Professor Sheena Lewis and her team have been developing this technique for over 20 years.
However, Prof. Lewis, who is emeritus professor of reproductive medicine at Queen’s University Belfast and chief executive of Examen, says that while developments in the use of AI are exciting, medicine moves very slowly.
For example, sperm detection is currently at the proof-of-concept stage, after a very small trial involving only seven patients.
“It still doesn’t make sense,” says Professor Lewis. “The time between something being proof-of-concept to being commercially available is usually between two and five years.
“It has a long way to go. It’s a target even for a very small group of men with NOA. It’s fantastic what you can do – but it’s never going to be mainstream.”
Back in Sydney, Dr Vasilescu says treatments like his are the “last stop”.
“It can be the difference between fertilizing the egg — or just stopping the treatment,” he says.
“If we can make embryologists more efficient, more accurate, they can find sperm that they otherwise wouldn’t. It gives a man a chance to father his own biological children.”
The UTS team is now ready to take their AI into clinical trials. “Real live pregnancy – that’s the next step,” says Dr Vasilescu.