An artificial intelligence has been developed by scientists to understand which ones are healthier by scanning the sperms.
Even in men who eat well and exercise regularly, only one in ten sperm is healthy, and in some men the rate drops below four percent. Scientists at the Oma Clinic in California have developed a “sperm health test” algorithm that screens sperm for their shape and how they move. They hope this study will increase success rates and reduce costs with IVF, meaning couples will need fewer treatment sessions.
Entrepreneur who helped develop AI, Dr. “Currently, couples have to go through an average of three IVF cycles,” Kiran Joshi told DailyMail.com. “We want to reduce this number as it will reduce the pain and financial burden and lead to more success.”
Studies show that about 40 percent of infertility cases are caused by the male partner, with sperm counts in decline for decades. Poor diet and increased sedentary lifestyles are cited as the reason for this trend.
Dr. Joshi explained in his interview that the algorithm was created by scanning sperm from the ejaculates of more than 1,000 men. Samples were submitted by fertility clinics in the US and other countries, and were mostly drawn from men in their 30s and 40s, but samples up to 75 years of age were also found.
To improve the algorithm, the scientists started by extracting a subset of sperm from each sample containing up to 20,000 sperm. Embryologists then placed the samples under the microscope and analyzed the health of each.
Sperm were graded based on two factors: their shape and how well they swim
The head of a healthy sperm is smooth and oval, and can swim quickly and in a straight line. However, Dr. Finding these good swimmers is often like “looking for a needle in a haystack,” Joshi says.
Sperm were graded according to these two aspects using standard measures, and then a video of about ten seconds of each was recorded. This data was then fed into artificial intelligence to give it the ability to accurately predict how healthy a particular sperm is.
As the AI works, it draws a box around each sperm and then quickly casts a color based on the sperm’s perceived health. Results are delivered in just seconds. Red boxes show unhealthy sperm, greens show healthy ones.
The scientists are preparing to publish their research in an academic journal in the next few months, although they have not disclosed any journal titles.