A new artificial intelligence (AI) tool has the potential to revolutionize the process of identifying viable sperm, offering hope to severely infertile men seeking to father biological children. Developed by researchers from the University of Technology Sydney, this algorithm can significantly improve the efficiency and accuracy of sperm identification, replacing the time-consuming manual methods used by embryologists.

Lead author Dale Goss explains that the AI tool has the capability to provide increased chances for patients with minimal prospects of fathering their own biological children. By streamlining the identification of sperm in samples, the algorithm not only enhances the chances of conception but also reduces stress on sperm and improves laboratory efficiency.

Severe infertility, known as non-obstructive azoospermia (NOA), affects approximately 1% of men and 5% of couples seeking fertility treatment. While these men may have minimal sperm quantities in their testes, the current procedure involves the removal of a tissue sample for manual examination by embryologists. This time-consuming process, taking up to six hours, can lead to mental and physical fatigue, potentially impacting the accuracy of sperm identification.

The AI tool, called SpermSearch, eliminates the need for this arduous process by conducting a rapid search for sperm in seconds. By training the algorithm using thousands of microscope photographs, the researchers enabled it to identify sperm among other cells and debris. In a test comparing the AI’s performance against an experienced embryologist, the AI tool demonstrated exceptional accuracy, identifying sperm in less than 1000th of the time and discovering more sperm overall.

The study’s findings, presented at the European Society of Human Reproduction and Embryology’s annual meeting, serve as a proof-of-concept, highlighting the need for further clinical trials. The development of this AI tool holds great promise in assisting clinicians and increasing the chances of conception for infertile men.

Disclaimer: The study mentioned in this article is based on a proof-of-concept test, and further research is necessary to validate the results and confirm the clinical application of the AI tool.

By Impact Lab