Fighting male infertility with AI

One in five couples in France faces infertility. In February, the government presented a plan to address infertility, including increasing the options for freezing eggs and better management of polycystic ovary syndrome, a common condition in women. However, difficulties in having children are not solely attributed to women. For men diagnosed with infertility, a new technique offers hope.

This technique, based on artificial intelligence, can detect sperm in men who were previously told they had none. About 10 to 15% of infertile men suffer from azoospermia, a condition where sperm is not present in semen samples under a microscope, even though there may be some. The new technique, called STAR (Sperm Track and Recovery), developed by Dr. Zev Williams at Columbia University in the United States, identifies and retrieves sperm, giving these men a chance to have biological children.

Researchers are using artificial intelligence to analyze hundreds of images per second, a task that would take technicians hours to complete. With this method, nearly 30% of cases have shown success in finding sperm in men who thought they had none. Ahmed Ziyyat from the Cochin Institute is part of the research team studying human reproductive disorders and sees promise in the STAR technique.

The technique has already helped a couple who had been trying to conceive for 19 years have a daughter. Since then, hundreds of people from around the world have sought assistance from Dr. Williams’ team. However, more large-scale clinical trials are needed to evaluate the long-term outcomes of this experimental technique.