HKUMed hails ‘world-first’ AI tool that picks best sperm with 96% accuracy, offering new hope for couples struggling with infertility

A research team from the Department of Obstetrics and Gynaecology at the University of Hong Kong's LKS Faculty of Medicine (HKUMed) has developed the world's first artificial intelligence (Al) model that can accurately identify human sperm with fertilisation potential. This breakthrough could reshape diagnosis and assisted reproductive treatments worldwide.

The Al model evaluates sperm morphology based on its ability to bind with the zona pellucida (ZP), which is the outer coat of the egg. By automating a process that has traditionally depended on manual and subjective analysis, the model has demonstrated a clinical validation accuracy rate exceeding 96%. This innovative approach outperforms traditional methods in terms of speed and reliability, reduces human error, and significantly enhances the precision of male fertility assessment-ultimately increasing the success rates of assisted reproductive procedures. The research findings were published in the international journal Human Reproduction Open [link to the publication] and won the Silver Award at the 50th Geneva International Invention Fair in 2025.

Infertility is a significant global health concern, affecting about one in six couples of reproductive age worldwide, with male factors accounting for 20–70% of cases. According to the World Health Organisation (WHO), infertility is projected to become the third most common disease globally, following cancer and cardiovascular diseases. While assisted reproduction treatments (ARTs) remain the most effective treatments for infertility, their success rates are limited by the accuracy of existing diagnostic tools.

The AI model invented by HKUMed’s Department of Obstetrics and Gynaecology in the School of Clinical Medicine has clinical value in evaluating male fertility. Conventional assessment methods rely heavily on subjective visual judgment and have inherent limitations, whereas the AI model precisely analyses subtle traits in sperm, enabling a more accurate prediction of fertilisation potential. The research team members include (from left) Dr Raymond Li Hang-wun, Professor Philip Chiu Chi-ngong, Professor William Yeung Shu-biu and Dr Erica Leung Tsz-ying.

Semen analysis is a standard clinical assessment for male fertility potential before ART. Traditionally performed manually under a microscope, this analysis assesses sperm morphology in accordance with WHO guidelines. However, Professor William Yeung Shu-biu, from the Department of Obstetrics & Gynaecology, School of Clinical Medicine, HKUMed, explained: 'This method is not only labour-intensive and time-consuming, but also highly dependent on the subjective judgment of laboratory technicians. This leads to significant variations between individuals and across laboratories, making it difficult to standardise sperm quality criteria and undermining the accuracy of male fertility evaluations.'

A typical male ejaculate contains 100 to 200 million motile sperm per millilitre, but only about 7% of these sperm have fertilisation potential. During natural conception, the selection mechanisms within the female reproductive tract eliminate inferior sperm, ensuring that only fertilisation-competent sperm can initiate fertilisation. However, ART laboratories currently lack an equally efficient sperm selection method and instead rely primarily on parameters from semen analysis such as sperm concentration, motility and morphology to guide fertilisation methods in ART, like in vitro fertilisation (IVF) and intracytoplasmic sperm injection (ICSI).

Professor Yeung explained: 'These traditional semen parameters have limitations in predicting the true fertilisation potential of male sperm. Even with normal semen analysis results, 5% to 25% of men still experience low fertilisation rates (less than 30%) or complete fertilisation failure during IVF. The failure in ART not only prolongs the time it takes for couples to conceive but also increases psychological stress and the financial burden.'

The AI model invented by HKUMed’s Department of Obstetrics and Gynaecology in the School of Clinical Medicine has clinical value in evaluating male fertility. Conventional assessment methods rely heavily on subjective visual judgment and have inherent limitations, whereas the AI model precisely analyses subtle traits in sperm, enabling a more accurate prediction of fertilisation potential. The research team members include (from left) Dr Raymond Li Hang-wun, Professor Philip Chiu Chi-ngong, Professor William Yeung Shu-biu and Dr Erica Leung Tsz-ying.

The binding of sperm to the ZP is the crucial first step in fertilisation. This layer selectively binds to sperm with normal morphology, intact chromosomes and fertilisation capability — a natural screening mechanism ensuring that only high-quality sperm fertilise the egg. Professor Philip Chiu Chi-ngong, Associate Professor in the same department and co-leader of the study, noted, 'Based on this physiological process, our team developed a highly automated Al model that analyses morphological features to accurately determine the percentage of human sperm capable of binding to the ZP, providing a highly reliable assessment of male fertility.'

The Al model developed by the HKU team is based on this selective binding mechanism and evaluates sperm quality from the egg's perspective, with a clinical threshold established at 4.9%. Men with less than 4.9% of sperm showing binding capability are considered at higher risk of fertilisation problems. ‘The Al model offers early warning of fertilisation issues and helps identify patients with impaired fertilisation in IVF,' Professor Chiu 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.'

HKUMed has developed the world-first AI-powered fertilisation-competent sperm identification tool, which accurately determines the percentage of human sperm capable of binding to the ZP, providing a highly reliable assessment of male fertility. The innovation won a Silver Award at the 50th Geneva International Invention Fair in 2025.

Using advanced deep-learning techniques, HKUMed researchers trained the Al model on more than 1,000 sperm images, achieving an accuracy rate over 96%. From 2022 to 2024, the team further validated the model by examining over 40,000 sperm images involving 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.

Professor Chiu highlighted the clinical value of Al in evaluating male fertility: 'Conventional assessment methods rely heavily on subjective visual judgment, which has inherent limitations. In contrast, our Al model precisely analyses subtle traits in sperm, enabling a more accurate prediction of fertilisation potential.'

For couples struggling with infertility, repeated attempts at ART are often required, inevitably leading to significant stress, disappointment and financial strain. HKUMed team is committed to seeking medical breakthroughs. This innovative technology reflects scientific progress and provides support for couples in need, helping them realise their reproductive dreams more quickly.

Professor Yeung added, 'The advent of Al allows us to assess sperm fertilisation capacity in a standardised, reproducible manner, improving clinical decision-making and enabling personalised treatment plans. This innovation has the potential to improve overall infertility management, reduce fertilisation failure rates, and shorten the time to pregnancy. We are currently conducting large-scale clinical trials to further validate the application of the Al model and hope to benefit more patients.'

A research team from the Department of Pharmacology and Pharmacy at the LKS Faculty of Medicine of the University of Hong Kong (HKUMed) has developed an innovative AI-based cardiovascular risk prediction tool, called CardiOmicScore. With a single blood test, the system can accurately forecast the future risk of six major cardiovascular diseases (CVDs): coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease and venous thromboembolism. It can also provide early warning signals up to 15 years before clinical onset. The findings were published in Nature Communications [link to the publication].

HKUMed develops a cardiovascular risk prediction tool that can accurately predict the future risk of six major cardiovascular diseases with a single blood test. The system can provide early warning signals up to 15 years before clinical onset. The research is led by Professor Zhang Qingpeng (left).

AI-based multiomics integration reflects the body’s real-time health status

CVDs remain the leading cause of death worldwide, accounting approximately 19.8 million fatalities in 2022 alone. In routine health assessments, physicians typically evaluate cardiovascular risk based on age, blood pressure, smoking and other conventional clinical indicators. However, these measures often fail to capture subtle and early biological changes before the disease becomes clinically apparent, leading to many patients missing the optimal window for preventive intervention. Although polygenic risk scores have become popular in recent years, genetic predisposition is largely fixed at birth and does not change over time. Consequently, polygenic risk scores cannot reflect the immediate impact on health conditions resulting from lifestyle or environmental changes. This creates an urgent need for tools that can capture a person’s current biological state and provide accurate, early warnings for CVDs.

To address this problem, the HKUMed research team applied deep learning techniques to integrate multiomics data, including genomics, metabolomics and proteomics, to develop the CardiOmicScore tool. The study was based on large-scale population data from the UK Biobank, analysing 2,920 circulating proteins and 168 metabolites measured from blood samples. These molecular signals act as ‘real-time recorders’ of the body, sensitively reflecting subtle changes in the immune system, metabolism, and vascular health.

Professor Zhang Qingpeng, Associate Professor in the Department of Pharmacology and Pharmacy at HKUMed, explained, ‘Genes determine where we start—they define our baseline health risk. However, proteins and metabolites reflect our current physical health. Our AI tool is designed to decode these complex molecular signals, enabling doctors and patients to identify risks much earlier, which can potentially change the trajectory of disease through timely lifestyle modifications and early prevention.’

Accurate prediction of six major cardiovascular diseases with 15-year advance warning in high-risk groups

The results showed that CardiOmicScore transforms complex multiomics measurements into personalised risk scores with substantially improved predictive performance compared with conventional polygenic risk scores. When combined with clinical information such as age and gender, the model significantly enhanced the risk prediction accuracy of six common CVDs and can even flag elevated risk up to 15 years before symptoms appear.

This study marks a shift in precision medicine from a static, gene-centric paradigm towards a more dynamic, multiomics-based approach. In the future, a small-volume blood sample may be sufficient to generate a comprehensive cardiovascular risk profile for multiple diseases.

Professor Zhang added, ‘We aim to leverage technology to identify and prevent diseases before they develop. By shifting health management from reactive treatment to proactive prediction and intervention, we aim to create a lasting impact for both public health and individual patient care.’

About the research team

The study was led by Professor Zhang Qingpeng, Associate Professor in the Department of Pharmacology and Pharmacy, HKUMed, and the HKU Musketeers Foundation Institute of Data Science (IDS). The first author is Luo Yan from the HKU IDS.