Researchers say an AI-modelled test for male infertility could soon be available to GPs
A new, accessible blood test that can predict male infertility could soon be used by GPs, according to researchers.
The study, published in the journal Scientific Reports, looked at data from nearly 4,000 men who underwent sperm and hormone testing to detect male infertility between 2011 and 2020.
From this, an AI model was developed that researchers say can predict the risk of male infertility with about 74% accuracy. It works by measuring different hormone levels in a blood sample, which are linked to sperm production.
The researchers said the model could predict a severe form of infertility known as non-obstructive azoospermia – where there is no sperm in the semen – with 100% accuracy.
The team believes their AI-enhanced blood test could make male infertility screening more accessible, as the test can be used in GP practices without the need for specialist laboratories.
Hideyuki Kobayashi, an associate professor at the department of urology at Toho University School of Medicine in Japan, who led the development of the AI model, said the method was only intended as a first screening step to identify infertility and was “not a replacement for sperm testing.”
He added: “In the future, we hope that clinical laboratories and health check-ups will use our AI prediction model to screen for male infertility, making male infertility testing more accessible by overcoming the barriers to it.”
According to the World Health Organization, approximately 7% of men worldwide suffer from infertility, with about half of fertility problems in heterosexual couples being male.
Allan Pacey, professor of andrology at the University of Manchester, said the method could help streamline the process of detecting male infertility.
He said: “One of the first steps in diagnosing male infertility is the analysis of a sperm sample in a specialist laboratory. This requires time off work and another appointment, sometimes in a specialist laboratory some distance away. Therefore, the idea that a first-stage diagnosis can be made from a blood sample taken by the GP does have some advantages.”
He added: “The authors of this paper have done a great job of applying artificial intelligence to the problem, but their approach should be simplified into some kind of app that GPs can use to benefit from it in practice.
“Clearly, the man will ultimately have to provide a semen sample for analysis, but if this approach is validated in a larger dataset, it could streamline the process and make it somewhat more user-friendly.”