A groundbreaking artificial intelligence (AI) approach has been developed by researchers at Columbia University, leading to the first successful pregnancy for a man who had battled infertility for nearly 20 years. This innovative technique, known as the STAR method, aims to address the difficulties associated with identifying and retrieving rare viable sperm cells in men with azoospermia, a condition impacting approximately 10-15% of infertile men.
The STAR method represents a significant advance in fertility treatment, particularly for those who have faced failed in vitro fertilization (IVF) cycles, manual searches for sperm, and surgical extractions without success. Traditional methods of recovering sperm from the testes often lead to complications such as vascular issues, inflammation, and decreased testosterone levels. Manual analysis of semen samples is time-consuming, expensive, and may damage sperm during processing.
“A semen sample can appear totally normal, but when you look under the microscope, you discover just a sea of cellular debris, with no sperm visible,” explained Zev Williams, the senior author of the study and Director of the Columbia University Fertility Center. “Many couples with male-factor infertility are told they have little chance of having a biological child,” he added.
The STAR method was designed to tackle the challenge of finding these rare sperm cells by employing advanced imaging, microfluidics, and robotics. The system utilizes high-resolution imaging technology, allowing it to scan a semen sample and capture over 8 million images in just one hour. AI algorithms rapidly analyze these images to detect viable sperm cells amidst the debris.
Moreover, the STAR method incorporates a microfluidic chip that utilizes tiny channels to isolate the specific portion of the semen containing the identified sperm cell. A robotic system follows this isolation, gently retrieving the sperm cell in milliseconds, ensuring it remains viable for embryo creation or future storage.
In its initial test, the STAR technology was applied to a couple undergoing infertility issues for two decades. The patient provided a 3.5 mL semen sample, which the STAR method scanned through approximately 2.5 million images over two hours. This process successfully identified and isolated two viable sperm cells, which were subsequently used to create two embryos, resulting in a successful pregnancy.
While this achievement is based on a single case, it highlights the potential of this AI-assisted technology to surmount long-standing challenges in treating male-factor infertility related to azoospermia. Researchers emphasize that “You only need one healthy sperm to create an embryo.” The effectiveness of the STAR method is currently being assessed through larger clinical studies, with findings published in The Lancet.
