Scientists from the Oma Clinic in California have developed an artificial intelligence (AI) algorithm that scans sperm to detect the healthiest ones for fertilisation, potentially increasing success rates for in vitro fertilisation (IVF) and reducing treatment costs. According to studies, around 40% of infertility cases are due to issues with male partners, with sperm counts falling in recent decades due to unhealthy diets and inactivity. The AI algorithm was built by scanning sperm from the ejaculates of over 1,000 men, mostly aged between 30 and 40 and submitted via fertility clinics across the United States and from other countries.
Initially, a subset of sperm was extracted from each sample, including up to 20,000 swimmers, which were then analysed under a microscope to evaluate their health. Sperm were graded according to two factors: their shape and how well they moved. A healthy sperm has an oval-shaped head and can swim in a straight line with speed. The evaluation was standardised, and about a ten-second video of each sperm was recorded. This data was fed into the AI to develop its ability to measure the health of any particular sperm accurately. During use, a box is drawn around each sperm, and its health status is rapidly assessed by a colour code. Healthy sperm gets a green box, while unhealthy sperm gets a red box.
Results are delivered within seconds, creating a potential revolution for the IVF industry by increasing fertilisation rates, significantly reducing treatment costs and making the journey to conception quicker and less emotionally burdensome. Statistics estimate that nearly 36,000 children are born annually in the United States with IVF treatment, accounting for about 1% of all US births. Experts predict that this figure will continue to rise as more couples delay starting a family.
The Oma Clinic scientists carried out tests that monitored the success rates of IVF rounds that used and those that did not use the new technology. Although researchers cannot yet reveal the results, they indicated that there was a significant difference between the two groups. An IVF embryologist was involved in developing the AI and noted that they found sperm with non-straight patterns, swimming in circles, or aberrant swimmers; providing an example of “finding a needle in the haystack.” The AI offers an accurate second opinion on healthy sperm that is difficult to determine by embryologists themselves.
Dr Kiran Joshi, an entrepreneur who helped develop the AI algorithm, commented that “couples have to go through three cycles of IVF on average…we want to reduce that number as that will reduce the pain and reduce the financial burden and lead to more success.” The technology offers the potential to alter the playing field, helping to advance embryo development and increase fertilisation rates. The researchers are set to publish their research in an academic journal over the next few months, although the name of the journal is still unrevealed.
