Comprehensive Effects of Magnesium Supplementation on Cardiometabolic Risk Factors: A Systematic Review and Dose-Response Meta-Analysis.

Polyendocrine metabolic ovarian syndrome (PMOS), formerly known as polycystic ovary syndrome (PCOS), represents the same clinical entity-characterized by hyperandrogenism, ovulatory dysfunction, and/or polycystic ovarian morphology-with the updated nomenclature reflecting an expanded understanding of its systemic metabolic nature rather than a distinct pathological condition (). And this condition is affecting approximately 10-13% of women globally (), with a prevalence rate as high as 10.01% in China (). The clinical presentation of PMOS is highly heterogeneous, primarily characterized by menstrual irregularities, hyperandrogenism, and polycystic ovarian morphology. This condition is not only strongly associated with adverse reproductive outcomes such as infertility and pregnancy complications (including gestational diabetes and preeclampsia) (–) but also is frequently complicated by comorbidities including obesity, insulin resistance, type 2 diabetes, cardiovascular diseases, and psychological disorders (–), posing significant threats to patients’ long-term quality of life. Early and accurate diagnosis of PMOS is crucial for implementing effective interventions and preventing potential complications (). However, due to regional population differences and the complexity of the disease, the diagnosis and management of PMOS remain persistent challenges in the fields of gynecology and endocrinology (). According to a World Health Organization (WHO) report, up to 70% of PMOS cases worldwide remain undiagnosed (), and 30% of women experience a diagnostic delay of more than two years (). Although the diagnostic and treatment criteria for PMOS were revised multiple times between 1990 and 2023 (, –), the current diagnostic process still faces several challenges. First, diagnosis relies on ultrasonography and detailed clinical and biochemical analyses-procedures that are time-consuming, labor-intensive, and require professional interpretation. Second, the diversity of clinical manifestations and the subjectivity of assessment may lead to diagnostic inconsistencies. Third, there is a lack of evaluation tools capable of integrating multidimensional information, making it difficult to achieve a comprehensive and precise description of the disease (). Therefore, how to achieve early identification and diagnosis of PMOS has become an important research focus in the field. In recent years, artificial intelligence (AI) technologies, particularly machine learning (ML) and deep learning (DL), have demonstrated significant potential in medical image analysis, disease prediction, and diagnostic assistance. AI models are capable of efficiently processing large-scale, multi-source complex data, automatically learning underlying patterns within them, thereby providing objective, reproducible, and precise diagnostic recommendations. This offers potential support in overcoming the limitations of manual diagnosis. ML and DL technologies have now been widely applied in the auxiliary diagnosis of various diseases, including lumbar spinal stenosis (), peripheral artery disease (), bipolar disorder (), chronic obstructive pulmonary disease (), Alzheimer’s disease (), etc. In the context of PMOS, AI applications primarily encompass prediction, diagnosis, classification, and screening for complications, showing promising prospects. Numerous studies have confirmed that in PMOS-related diagnostic tasks, ML or DL models not only exhibit strong diagnostic performance but have even matched or surpassed the level of clinical experts (). For instance, Zhao et al. developed a DL-based model for the automated recognition of polycystic ovary ultrasound images. The results showed that the model achieved area under the curve (AUC) values of 0.953, 0.973, and 0.967 in the training set, internal validation set, and external validation set, respectively (). Moreover, the model evaluated a single ovary 50 times faster than clinicians, underscoring the significant value of AI tools in the diagnosis of PMOS ().