A new algorithm can determine male fertility at a faster and more accurate rate than previously possible, according to research by UK startup Bayezian.
The breakthrough arrives amid growing issues for couples trying to get pregnant.
A recent report from the World Health Organisation estimated that one in six people globally is now affected by infertility. Despite perceptions that it’s “women’s business,” men now contribute to approximately 50% of fertility problems.
Indeed, the male factor has become a growing concern. Recent research found that sperm counts have dropped by more than 50% over the past 45 years, with double the rate of decline since 2000. Up to 7% of men are now affected by infertility, but getting a diagnosis can be slow, expensive, and inconclusive.
These issues have triggered calls for better fertility testing. Around 18 months ago, Bayezian was asked to help. The company, which provides data science and machine learning incubation services, applied AI to the problem.
“We see accurate diagnosis as a critical tool in helping address male fertility.
Bayezian sought a solution in the MHSMA dataset, a collection of sperm images from 235 patients with male factor infertility. Each image is labelled by experts for normal or abnormal sperm acrosome, head, vacuole, and tail, which has made it an attractive dataset for machine learning studies.
Using the dataset, the research team built deep learning frameworks that can see a sperm’s morphology.
According to Bayezian, their algorithm spots differences that the human eye can’t perceive. The company says it can identify sperm fertility with a 96% accuracy rate — 2% higher than existing scientific approaches.
“This project is the perfect example of the tech for good approach that the team is undertaking,” said Ed Dixon, founder and CEO of Bayezian. “We see accurate diagnosis as a critical tool in helping address male fertility.”