TLDR: A research team at HKUMed has developed the world’s first AI model to accurately identify human sperm with high fertilization potential, achieving over 96% accuracy. This innovation aims to revolutionize male fertility diagnosis and assisted reproductive treatments globally, addressing a significant cause of infertility.
A pioneering research team from the Department of Obstetrics and Gynaecology at the University of Hong Kong’s LKS Faculty of Medicine (HKUMed) has announced the development of the world’s first artificial intelligence (AI) model capable of precisely identifying human sperm with fertilization potential. This breakthrough, published on August 25, 2025, promises to significantly advance diagnosis and assisted reproductive treatments worldwide, offering new hope to couples grappling with infertility.
The AI model operates by evaluating sperm morphology based on its ability to bind with the zona pellucida (ZP), the outer layer of the egg. Traditionally, this assessment has relied on manual and subjective analysis, which is prone to human error and limitations. The new AI-driven approach automates this process, demonstrating a clinical validation accuracy rate exceeding 96%. This remarkable precision and speed represent a substantial improvement over conventional methods, enhancing the reliability of male fertility assessment and, consequently, the success rates of assisted reproductive procedures.
Infertility remains a pressing global health issue, affecting approximately one in six couples of reproductive age. Male factors contribute to a significant portion of these cases, ranging from 20% to 70%. The World Health Organisation (WHO) projects infertility to become the third most common disease globally, following cancer and cardiovascular diseases. While assisted reproductive treatments (ARTs) are effective, their success has been constrained by the accuracy of existing diagnostic tools.
The HKUMed researchers utilized advanced deep-learning techniques to train their AI model on over 1,000 sperm images. Between 2022 and 2024, the model underwent further validation by examining more than 40,000 sperm images from 117 men diagnosed with infertility or unexplained infertility. The results confirmed a strong correlation between the proportion of sperm capable of binding to the ZP and the success rate of ART procedures.
Also Read:
- Generative AI Identifies Hidden Bird Flu Exposure Risks in Maryland Emergency Departments
- China’s AI Future: 36Kr’s 2025 Industry Conference to Unveil “Chinese Solution”
Professor Chiu, a member of the research team, stated, ‘The AI model offers early warning of fertilisation issues and helps identify patients with impaired fertilisation in IVF.’ He added, ‘It serves as a novel diagnostic tool for detecting fertility issues that conventional semen analysis may overlook, allowing clinicians to tailor more effective treatment plans and improve pregnancy outcomes.’ This innovative tool accurately determines the percentage of human sperm capable of binding to the ZP, providing a highly reliable assessment of male fertility. The research findings were published in the international journal Human Reproduction Open and received the Silver Award at the 50th Geneva International Invention Fair in 2025.
