Age-related Curves of AMH Using the Gen II, the picoAMH, and the Elecsys Assays in Women With Polycystic Ovary Syndrome

SmartScanPCOS: A feature-driven approach to cutting-edge prediction of Polycystic Ovary Syndrome using Machine Learning and Explainable Artificial Intelligence

PolyCystic Ovarian Syndrome (PCOS) poses significant challenges to women's reproductive health due to its diagnostic complexity arising from a variety of symptoms, including hirsutism, anovulation, pain, obesity, hyperandrogenism, and oligomenorrhea, necessitating multiple clinical tests. Leveraging Artificial Intelligence (AI) in healthcare offers several benefits that can significantly impact patient care, streamline operations, and improve medical outcomes overall. This study presents an Explainable Artificial Intelligence (XAI)-driven PCOS smart predictor, structured as a hierarchical ensemble consisting of two tiers of Random Forest classifiers following extensive analysis of seven conventional classifiers and two additional stacking ensemble classifiers. An open-source data set comprising numerical parametric features linked to PCOS for classifier training was used. Moreover, to identify essential features for PCOS prediction three feature selection methods: Threshold-driven Optimized Principal Component Analysis (TOPCA), Optimized Salp Swarm (OSSM), and Threshold-driven Optimized Mutual Information Method (TOMIM) were fine-tuned through thresholding and improvisation to detect diverse attribute sets with varying numbers and combinations. Notably, the two-level Random Forest classifier model outperformed others with a remarkable 99.31 % accuracy by employing the top 17 features selected through the Threshold-driven Optimized Mutual Information Method (TOMIM) along with anoverallaccuracy of 99.32 % with 8 fold cross validation for 25 runs. The Smart predictor, constructed using Shapash - a Python library for Explainable Artificial Intelligence - was utilized to deploy the two-level Random Forest classifier model. Ensuring transparency and result reliability, visualizations from robust Explainable AI libraries were employed at different prediction stages for all considered classifiers in this study.

Age-related Curves of AMH Using the Gen II, the picoAMH, and the Elecsys Assays in Women With Polycystic Ovary Syndrome

CONTEXT

Several challenges still exist to adopt the anti-müllerian hormone (AMH) as a marker of polycystic ovary morphology, as included in the recently updated international guideline. Although different evaluations of age- and assay-specific reference ranges have been published in the past few years, these studies have mainly been conducted in normo-ovulatory or infertile women.

OBJECTIVE

To develop an age-specific percentile distribution of AMH in patients with polycystic ovary syndrome (PCOS) measured by 3 different assays.

DESIGN

Retrospective cross-sectional study.

PATIENTS

A total of 2725 women aged 20 to 40 years with PCOS diagnosis were included.

INTERVENTIONS

Serum AMH measurement by the Gen II (Beckman Coulter), the picoAMH (Ansh Labs), and the Elecsys (Roche) assays.

MAIN OUTCOME MEASURES

Age-specific percentile curves for all the assays and correlations between AMH, clinical, hormonal, and ultrasound characteristics.

RESULTS

Age-related nomograms for the 5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles of AMH were calculated using the Lambda-Mu-Sigma method for all the assays. AMH levels were significantly different between PCOS phenotypes. AMH levels were positively correlated to LH, LH/FSH ratio, testosterone, androstenedione, free androgen index, mean follicular number, and mean ovarian volume.

CONCLUSION

To our knowledge, this is the first study reporting age-specific percentile nomograms of serum AMH levels measured by the Gen II, the picoAMH, and the Elecsys assays in a large population of women with PCOS. These findings may help to interpret AMH levels in patients with PCOS and facilitate the use of AMH as a diagnostic tool across age ranges.

Anti-Müllerian hormone a surrogate of follicular fluid oxidative stress in polycystic ovary syndrome?

Polycystic ovary syndrome (PCOS) is the most common endocrinopathy in women at childbearing age. Anti-Müllerian hormone (AMH) is a widely accepted sensitive marker of ovarian reserve, which has been suggested that could also act as biomarker of ovarian morphology for PCOS diagnosis. Oxidative stress (OS) is known to be associated and have a negative impact factor in several reproductive conditions, including PCOS. However, the relationship between circulating AMH and OS within the follicular fluid (FF), and its potential impact on in vitro fertilization (IVF) outcomes of women with PCOS, remains largely unexplored. A total of 84 women, with PCOS (n = 30) or ovulatory controls (n = 54), were enrolled in this study. Women underwent individualized controlled ovarian stimulation for oocyte retrieval. Blood and FF obtained from mature follicles were collected at the time of oocyte retrieval, for measuring total testosterone, ∆4-androstenedione, progesterone, sex hormone binding globulin (SHBG) and AMH. OS in the FF was assessed by measuring total antioxidant capacity (TAC) through the ferric reducing antioxidant power (FRAP) and lipid peroxidation (LPO) by quantification of malondialdehyde (MDA) levels. Our results demonstrated that women with PCOS had significantly higher plasma levels of AMH, ∆4-androstenedione, total testosterone and a free androgen index (FAI) than observed in non-PCOS controls. In women with PCOS, total testosterone and AMH levels in the FF were also higher, while TAC was lower compared to non-PCOS. Furthermore, circulating AMH levels were positively correlated with ∆4-androstenedione, albeit negatively correlated with TAC. In this study we demonstrated that the susceptibility to OS, as assessed by the total antioxidant capacity in the FF, is higher in women with PCOS and inversely related to AMH levels. This study results lead us to forge the reasonable hypothesis that the greater susceptibility to OS within the follicle microenvironment is potentially at the end of a roadway that starts with elevated ∆4-androstenedione and AMH within the FF, which in turn are mirrored by circulating AMH and androgen levels. Thus, suggesting that circulating AMH levels could act as a surrogate biomarker of follicular fluid oxidative stress in women with PCOS.