The review, led by Li Tang and colleagues at Peking University Third Hospital, is a structured narrative synthesis rather than a single experiment. The team searched PubMed and Web of Science for literature published between 2013 and 2025, screening roughly 400 records and ultimately including 68 articles that focus on machine learning and deep learning models applied to biochemical and imaging data in ovarian reserve evaluation. By systematically mapping how these algorithms have been used—what inputs they consume, how they are trained, and how they perform—the authors have produced one of the clearest portraits to date of a field moving rapidly from proof-of-concept studies toward tools that could reshape reproductive medicine.

At the technical heart of the review lies a distinction between what conventional diagnostics measure and what machine learning can infer. Traditional biomarkers such as AMH are produced by small growing follicles and correlate broadly with the number of resting follicles, but they say little about follicular quality, the trajectory of decline in an individual woman, or how she might respond to ovarian stimulation during assisted reproductive technology (ART). Imaging offers complementary information: antral follicle counts, ovarian volume, and stromal characteristics visible on ultrasound carry predictive signal that blood tests miss. The problem has historically been that these data streams live in separate silos, interpreted by different specialists using subjective visual criteria. Machine learning models, and particularly multimodal architectures that fuse inputs from multiple sources, are designed precisely to dissolve those silos.

One of the most mature applications documented in the review is automated follicle detection on ultrasound. Deep convolutional neural networks—the same class of models that powers facial recognition and autonomous driving—can be trained on annotated ultrasound images to identify and count antral follicles with a consistency that human observers struggle to match across scans, machines, and sonographers. Inter-observer variability in manual follicle counting is a well-known source of error in ovarian reserve testing; a follicle near the size threshold may be included by one operator and excluded by another. Automated segmentation and detection pipelines reduce this noise while dramatically speeding up the analysis, and the review reports that representative AI systems in this domain have achieved sensitivity of approximately 90 percent and specificity of 80 percent in classifying ovarian reserve status—performance figures that approach or exceed those of conventional workflows.

Beyond counting follicles, the review highlights how AI models can perform genuine multimodal fusion. Rather than treating AMH, FSH, estradiol, and ultrasound features as parallel numbers to be eyeballed, modern architectures such as multi-input neural networks learn cross-modal relationships: how a given hormone level changes its predictive meaning depending on follicular morphology, age, menstrual cycle characteristics, or clinical history. Some models described in the literature ingest longitudinal data, tracking how a patient’s markers evolve across cycles to estimate the individual rate of ovarian aging. The authors emphasize that this capacity for integration enables what they call personalized “endocrine age” modeling—an estimate of how fast a particular woman’s reproductive endocrine system is aging, which may diverge substantially from her chronological age.

Natural language processing adds a third, less obvious data stream. Decades of clinical narratives—consultation notes, procedure records, and cycle outcome descriptions stored in electronic health records—contain rich information about ovarian response that is locked in free text. Transformer-based language models can extract structured variables from these documents at scale, allowing researchers to assemble training datasets far larger than what manual chart review would permit. Combined with genetic data, which the review notes can flag conditions such as fragile X-associated diminished ovarian reserve, these extracted variables expand the feature space available to predictive models of ART outcomes, including ovarian stimulation response and oocyte yield.

The clinical implications are substantial. In fertility clinics, ovarian reserve testing drives decisions about the timing of treatment, the choice and dosing of gonadotropin stimulation protocols, and the counseling of women considering delaying childbearing. Inaccurate assessments carry real costs: a patient misclassified as a poor responder may be subjected to unnecessary anxiety or inappropriate high-dose stimulation, while a falsely reassuring profile can delay treatment until options have narrowed. By improving both the accuracy and the granularity of assessment, multimodal AI could support more individualized stimulation protocols, better prediction of hyper-response risk in conditions such as polycystic ovary syndrome, and more honest, data-grounded fertility counseling. The review frames this as a movement from isolated biomarkers toward personalized, precise reproductive care.

Yet the authors are careful to temper enthusiasm with a sober accounting of the obstacles standing between published algorithms and routine clinical use. Data standardization is the most immediate challenge. Ultrasound images acquired on different machines with different settings, hormone assays with different calibration standards, and clinical records encoded in incompatible formats all degrade the ability of models to generalize. Models trained on data from a single center routinely perform worse when deployed elsewhere, and the review stresses that large-scale, multicenter validation is essential before any AI-driven ovarian reserve tool can be trusted across diverse patient populations. In a domain as consequential as fertility, a model that works only in the population where it was trained is not a clinical tool but a local artifact.

Explainability presents a second, equally serious barrier. Deep learning models are notoriously opaque: a network may output a risk score without articulating which features drove the prediction. Clinicians asked to base stimulation decisions or counseling conversations on an algorithm’s output need to understand why the algorithm reached its conclusion—both to exercise appropriate judgment and to earn patient trust. The review notes that developing clinician-oriented decision-support tools, in which AI outputs are presented in interpretable form alongside traditional measures rather than replacing them, remains an unfinished task. The authors also point to broader issues of external validation, regulatory oversight, and integration into existing clinical workflows as prerequisites for real-world adoption.

Taken together, the review sketches a field at an inflection point. The technical ingredients—automated image analysis, multimodal fusion architectures, language models capable of mining clinical text, and growing repositories of longitudinal reproductive data—have matured to the point where reported performance is clinically meaningful. What remains is the slower, harder work of medicine: harmonizing data across institutions, validating models in the populations they will serve, making their reasoning legible to the clinicians who must act on them, and embedding them into the daily rhythms of fertility care. If that work succeeds, the authors conclude, the assessment of ovarian reserve may shift from a snapshot of scattered numbers toward a dynamic, individualized portrait of reproductive aging—one in which hormones, images, and clinical histories are finally read together rather than apart. For the millions of women who undergo fertility testing each year, that integration could translate into better-timed treatments, safer stimulation protocols, and answers that are more precise than any single blood test or ultrasound has ever been able to provide.

Subject of Research: Artificial intelligence for multimodal assessment of ovarian reserve

Article Title: Bridging hormones and imaging: artificial intelligence advances multimodal quantitative and qualitative assessment of ovarian reserve

Article References: Tang, L., Ma, Z., Li, Y., Jin, Z., Zhang, J., Hu, H., Jin, Y., Lin, R., Chai, X., Wang, L., Heng, D., Zhang, W., Zhang, H., Xin, X., Zhu, Y., Ye, Y., Li, J., & Li, D. (2026). Bridging hormones and imaging: artificial intelligence advances multimodal quantitative and qualitative assessment of ovarian reserve. Journal of Ovarian Research. https://doi.org/10.1186/s13048-026-02245-0

Image Credits: AI Generated

DOI: 10.1186/s13048-026-02245-0

Keywords: ovarian reserve, artificial intelligence, machine learning, deep learning, anti-Müllerian hormone, ultrasound, follicle detection, multimodal fusion, fertility, assisted reproductive technology, endocrine age, reproductive medicine

The team’s research centers on two primary dimensions: quantitative and qualitative assessment of the ovarian reserve. Traditionally, medical professionals have relied on hormonal levels and imaging techniques to assess ovarian function. However, these measures often provide an incomplete picture, lacking the depth and precision necessary for tailored interventions. Koike and his colleagues aim to enrich this understanding by harnessing the predictive capabilities inherent in machine learning algorithms. By mining vast datasets, the researchers hope to uncover hidden patterns that can lead to more nuanced evaluations of ovarian health.

Artificial intelligence, particularly machine learning, serves as the backbone of this innovative approach. Unlike traditional statistical methods, machine learning algorithms can process large volumes of data, making determinations based on non-linear relationships and complex interactions that might elude human analysis. The researchers meticulously gathered demographic data, hormonal profiles, and ultrasound images from a diverse cohort of women. This multi-faceted dataset is critical for training the algorithms to make accurate predictions regarding ovarian reserve.

One of the key advantages of machine learning lies in its adaptability. Models can continuously improve as they are exposed to new data, a feature that could refine assessments of ovarian reserve over time. As more women seek fertility treatments or take proactive steps to manage their reproductive health, validated models could serve as essential tools for clinicians. The ability to predict outcomes with higher accuracy may shift the paradigm toward preventive measures, reducing the emotional and financial toll often associated with fertility challenges.

A particularly compelling aspect of the study involves the qualitative analysis of ovarian reserve, which includes examining the health and viability of oocytes. This dimension is crucial as it can profoundly affect fertility outcomes. Past methodologies did not adequately address how the quality of oocytes changes in relation to the ovarian reserve. By employing sophisticated machine learning techniques, the researchers aim not only to predict how many follicles will respond but also to assess their potential for successful fertilization.

Furthermore, the implications of this research extend beyond the laboratory. As fertility specialists begin to adopt these new assessment tools, the potential for personalized medicine becomes a reality. Women could receive data-driven insights tailored to their unique circumstances, guiding them through their reproductive choices with greater confidence. This bespoke approach may include recommendations on lifestyle modifications, timing for conception attempts, or interventions that could enhance fertility prospects.

Another exciting angle of the research is its potential application in clinical settings. If validated through rigorous testing, this machine learning-enabled model could become a routine part of reproductive health assessments. A more accurate understanding of ovarian reserve could lead to earlier interventions, potentially improving outcomes for women at risk of diminished ovarian function. Such proactive measures can significantly alter the landscape of reproductive health, challenging existing norms and reshaping patient care protocols.

The study also touches on ethical considerations surrounding the use of artificial intelligence in medicine. As algorithms begin to inform critical health decisions, questions arise regarding data privacy, informed consent, and the risk of bias in machine-driven recommendations. Koike and colleagues have proactively addressed these issues by emphasizing the importance of transparent algorithms and the need for continual oversight by medical professionals. Such vigilance is essential to maintain patient trust as we navigate the complexities of AI in healthcare.

Importantly, the research underlines the urgency of advancing our understanding of ovarian health amidst increasing infertility rates globally. Economic factors, lifestyle changes, and environmental influences are converging to create challenges for current and future generations of women. By investing in research such as this, the medical community can begin to formulate more effective strategies to tackle these pressing challenges. The incorporation of technology into reproductive health is not merely a trend; it’s a necessary evolution.

In conclusion, the study led by Koike et al. represents a significant leap forward in the quest to better understand ovarian reserve through machine learning. The interplay between artificial intelligence and reproductive medicine holds untold potential, promising a future where women can make informed reproductive choices based on precise data. As the findings of this research ripple through the healthcare landscape, we may witness a shift not only in clinical practice but also in the broader discourse surrounding women’s health, fertility, and the integration of technology in personal wellness.

This remarkable endeavor epitomizes the power of interdisciplinary collaboration, where cutting-edge technology and a commitment to women’s health converge. As researchers continue to refine their models and validate their findings, the medical community and the women they serve stand on the precipice of a new era, poised to embrace the possibilities that lie ahead. In the coming years, monitoring advancements in this field will be crucial, as the integration of machine learning into reproductive health could reshape the future of fertility medicine and empower women to take charge of their reproductive journeys.

Subject of Research: Assessment and prediction models for the ovarian reserve using machine learning.

Article Title: Assessment and prediction models for the quantitative and qualitative reserve of the ovary using machine learning.

Article References:

Koike, H., Harada, M., Yoshida, K. et al. Assessment and prediction models for the quantitative and qualitative reserve of the ovary using machine learning.
J Ovarian Res 18, 153 (2025). https://doi.org/10.1186/s13048-025-01732-0

Image Credits: AI Generated

DOI: 10.1186/s13048-025-01732-0

Keywords: ovarian reserve, machine learning, fertility, reproductive health, artificial intelligence.

The study was unveiled at the 41st Annual Meeting of the European Society of Human Reproduction and Embryology (ESHRE), one of the foremost global platforms for cutting-edge reproductive science. This extensive population-based research was orchestrated by the EU FEMaLe consortium (Finding Endometriosis through Machine Learning), led by Dr. Lucky Saraswat of the University of Aberdeen, in collaboration with specialists at the University of Edinburgh. By mining an unprecedented dataset that integrates primary care, secondary care, and maternity records over a 30-year span—from 1991 to 2020—the team conducted the largest and most comprehensive analysis addressing the intersection of infertility and endometriosis to date.

Endometriosis is a chronic gynecological disorder characterized by the aberrant growth of endometrial-like tissue outside the uterine cavity, frequently precipitating debilitating pelvic pain and compromised fertility. Worldwide, it affects approximately 190 million women and is a leading cause of female infertility. Despite its high prevalence, endometriosis remains notoriously underdiagnosed, in part due to nonspecific symptoms and diagnostic delays. This study’s scope and duration have enabled a rare examination of the natural history and reproductive trajectories of affected women with an unparalleled level of granularity and clinical relevance.

The cohort encompassed 4,041,770 women aged between 13 and 50 years who sought medical attention in England for infertility or symptoms potentially attributable to endometriosis. Of these, 245,994 presented explicitly with infertility issues, while 111,197 had surgically confirmed endometriosis diagnoses, established via definitive procedures such as laparoscopy or laparotomy. The overall prevalence of female infertility stood at 48.9 per 1,000 women across the study period, peaking among those between 30 to 39 years old—an age range critical to reproductive potential and clinical decision-making.

Intriguingly, among the women identified with infertility, 6.1% (14,904) had a surgical diagnosis of endometriosis. Notably, 57.4% of this subgroup experienced infertility prior to receiving their endometriosis diagnosis, underscoring the persistent challenge of delayed recognition even in symptomatic individuals. This temporal disconnect highlights the urgent need for improved diagnostic tools and clinical vigilance to expedite identification and management of endometriosis in patients presenting with fertility disruptions.

Statistical analysis reaffirmed a robust association between endometriosis and infertility: women grappling with infertility were found to be more than twice as likely to have endometriosis diagnosed compared to those without fertility concerns. This correlation aligns with longstanding clinical observations, reinforcing the role of endometriosis in disrupting normal reproductive physiology through mechanisms such as impaired ovulation, pelvic adhesion formation, and altered immune and inflammatory environments.

Yet, when dissecting pregnancy outcomes among women with infertility, a counterintuitive yet compelling revelation emerged. Those whose infertility was linked to endometriosis demonstrated a fourfold greater likelihood of achieving conception relative to women whose infertility stemmed from other causes—be it ovulatory dysfunction, tubal pathology, or unexplained origins. Overall, 40.5% of women with an endometriosis diagnosis achieved at least one pregnancy during the longitudinal follow-up, a figure that challenges generalized perceptions of inevitable infertility associated with the disease.

Dr. Saraswat elaborated on these surprising findings by suggesting variability in disease phenotype and severity as a critical factor influencing reproductive potential. Women with milder manifestations of endometriosis may retain robust fertility, especially if the condition is identified and managed proactively. Surgical intervention, particularly laparoscopic excision or ablation, has demonstrated moderate-quality evidence of enhancing pregnancy rates, possibly through restoration of pelvic anatomy and reduction of localized inflammation.

Furthermore, heightened awareness among women diagnosed with endometriosis about the reproductive implications of their condition may prompt earlier and more assertive fertility consultations and treatments, thereby improving conception likelihood. This underscores the importance of timely diagnosis and patient education in optimizing fertility outcomes, as well as the multifactorial nature of reproductive success that encompasses age, disease burden, and treatment modalities.

The study’s robust dataset and longitudinal design enable a nuanced understanding of infertility trajectories in the context of endometriosis, delivering invaluable evidence to inform clinical counseling. Women newly diagnosed with endometriosis can now receive more precise prognostic information relating to the probability of infertility, the potential for successful conception, and how these outcomes compare with other infertility etiologies. Such evidence-based guidance is essential to empower patients in reproductive decision-making and to tailor therapeutic strategies to individual risk profiles and aspirations.

Beyond immediate clinical implications, these insights pave the way for future research endeavors to elucidate the interplay between disease stage, anatomical site of lesions, surgical management, and assisted reproductive technologies in shaping fertility outcomes. Understanding the heterogeneity of endometriosis and its impact on reproductive biology at a molecular and systemic level remains a critical frontier in reproductive medicine.

Professor Dr. Anis Feki, Chair-Elect of ESHRE, emphasized the study’s significance, highlighting how it offers reassurance to patients burdened by endometriosis-related infertility and reinforces the imperatives of early diagnosis and personalized treatment strategies. This research contributes to dismantling the stigma and fatalism often associated with endometriosis, fostering a more hopeful and scientifically grounded narrative for affected women.

The findings have been submitted for publication in Human Reproduction, a premier journal in reproductive biology and medicine, ensuring dissemination to global scientific and clinical communities. This study exemplifies the transformative potential of big data and machine learning methodologies in unraveling complex bio-clinical phenomena, heralding a new era of precision reproductive healthcare.

In sum, this landmark investigation not only validates the link between endometriosis and infertility but crucially delineates the nuanced reproductive prospects for these women. It calls for enhanced diagnostic acumen, individualized clinical pathways, and sustained research investment to unravel the intricacies of endometriosis and optimize fertility outcomes, transforming the reproductive landscape for millions worldwide.

Subject of Research: Reproductive outcomes and fertility potential in women with endometriosis-associated infertility based on a 30-year national population study.

Article Title: Not specified in the source content.

News Publication Date: 2 July 2025

Web References:
https://www.who.int/news-room/fact-sheets/detail/endometriosis#:~:text=Overview,period%20and%20last%20until%20menopause

References:
[1] Saraswat, L., et al. (2025). Infertility and endometriosis: a 30-year long national population-based study of prevalence, association and pregnancy outcomes. Human Reproduction.
[2] Rogers, P. A.W., et al. (2009). Priorities for endometriosis research: recommendations from an international consensus workshop. Reproductive sciences (Thousand Oaks, Calif.), 16(4), 335–346.
[3] World Health Organization. (2023). Endometriosis.

Image Credits: Not provided.

Keywords: Endometriosis, Infertility, Reproductive biology, Gynecological disorders, Fertility outcomes, Population-based study, Epidemiology, Surgical diagnosis, Laparoscopy, Assisted reproduction, Women’s health.