AI Blood Test Predicts Male Infertility Without Semen Analysis

A groundbreaking study from Japan has developed a way to predict male infertility using only a blood test, removing the need for traditional semen analysis. The research, led by Associate Professor Hideyuki Kobayashi from the Department of Urology at Toho University School of Medicine in Tokyo, was published in the British journal Scientific Reports.

According to a 2017 World Health Organization (WHO) report, male factors are responsible for about half of all infertility cases. Semen analysis is essential for diagnosing male infertility, but it is often only available at specialized clinics, making access difficult for many men.

To address this challenge, researchers created an artificial intelligence (AI) model that analyzes hormone levels in the blood to predict infertility risk. The model was developed using AI software that does not require any programming knowledge. It was trained on data from 3,662 men who had undergone both semen and hormone testing between 2011 and 2020.

The model showed promising results, predicting male infertility with about 74% accuracy. Remarkably, it was 100% accurate in identifying non-obstructive azoospermia — a severe form of infertility where no sperm is present due to a failure in sperm production.

The semen tests in the study measured volume, sperm concentration, and motility. Blood tests measured levels of hormones such as LH, FSH, PRL, testosterone, and E2, along with the testosterone-to-E2 (T/E2) ratio. From the semen data, researchers also calculated the total number of motile sperm, a key indicator of fertility.

The AI model used WHO’s 6th edition (2021) reference values for semen analysis to categorize fertility status and train its predictions.

This new method offers a more accessible and less invasive way to screen for male infertility. It could help men seek timely treatment without the discomfort or embarrassment often associated with semen testing.