Infertility affects millions of couples in the United States, with male infertility accounting for a significant portion of cases. In fact, Cleveland Clinic reports that 10% of all males in the country attempting to conceive suffer from infertility.
Non-obstructive azoospermia (NOA), a severe form of male infertility characterized by the absence of sperm in the semen, usually poses this challenge to couples.
InterestingEngineering reports that researchers at the University of Technology Sydney have developed an artificial intelligence (AI) tool called SpermSearch that shows promising results in identifying sperm in severely infertile men.
What Is Non-obstructive Azoospermia (NOA)?
NOA is a condition where no sperm are present in the semen, making it one of the most severe forms of male infertility.
Approximately one percent of all males and five percent of couples seeking fertility treatments are affected by NOA.
Currently, men with NOA undergo invasive procedures such as testicular sperm aspiration (TESA) or testicular sperm extraction (TESE) to retrieve viable sperm from their testes for use in fertility treatments like Intracytoplasmic Sperm Injection (ICSI).
SpermSearch
To avoid these invasive procedures altogether, scientists at the University of Technology Sydney have developed an AI tool called SpermSearch to overcome the challenges associated with identifying sperm in severely infertile men.
The algorithm, presented at the European Society of Human Reproduction and Embryology, utilizes machine learning and thousands of microscope photos featuring sperm, cells, and debris as training data.
The team trained the AI algorithm to recognize and differentiate between sperm and surrounding elements.
Read Also: Researchers Warn 'Disease X' Could Be Next Pandemic: What's This Pathogen?
To validate the tool, testicular tissue samples were collected from seven patients diagnosed with NOA who had previously undergone surgical sperm removal.
The AI algorithm and an experienced embryologist examined the sperm samples simultaneously, with the embryologist's identification as the benchmark for comparison.
Findings and Potential Impact
The AI tool outperformed the embryologist in terms of both times taken to identify sperm and accuracy.
The algorithm identified 611 sperm, compared to the embryologist's 560, showcasing its efficiency in identifying sperm within samples.
Moreover, SpermSearch demonstrated greater accuracy, identifying 60 more sperm and achieving five percent higher precision per viewable droplet area than the embryologist.
The potential impact of SpermSearch is significant. It offers a rapid and accurate method of identifying sperm, reducing the time and effort required by embryologists during fertilization procedures.
By streamlining the process, SpermSearch can enhance efficiency in the laboratory, increase success rates, and alleviate mental and physical exhaustion experienced by embryologists.
Future Research
While the initial results are promising, further clinical trials are necessary to test the effectiveness and usefulness of the AI tool. The researchers suggest expanding the testing to include men with other forms of infertility who undergo surgical procedures.
Lead author Dale Goss emphasizes the potential of SpermSearch to provide hope for couples with limited chances of conceiving their own biological children.
By improving upon outdated approaches, this AI tool has the potential to increase efficiency, reduce stress on sperm, and ultimately improve the chances of successful fertility treatments.
Stau posted here at Tech Times.
Related Article: Groundbreaking Brain Procedure Inside Womb Saves Baby's Life
