EC Gynaecology

Research Article Volume 15 Issue 10 - 2026

Incremental Discrimination of Clinical, Biochemical, and Ultrasound Features for Polyendocrine Metabolic Ovarian Syndrome (Formerly PCOS): A Secondary Analysis of a Public Multicenter Dataset

Mariam A Alkhatib1, Ahed J Alkhatib2* and Nawal A Alkhatib3

1Faculty of Nursing, Jordan University of Science and Technology, Jordan
2Department of Legal Medicine, Toxicology and Forensic Medicine, Jordan University of Science and Technology, Jordan
3Faculty of Pharmacy, Jordan University of Science and Technology, Jordan

*Corresponding Author: Ahed J Alkhatib, Department of Legal Medicine, Toxicology and Forensic Medicine, Jordan University of Science and Technology, Jordan
Received: September 17, 2026; Published: October 07, 2026



Background: PMOS is diagnosed with combinations of reproductive, androgenic and ovarian features; it was previously known as polycystic ovary syndrome (PCOS). The public datasets on PCOS have been used widely for machine-learning classification. However, not much attention has been paid to the extra information offered by the clinical, biochemical, and ultrasound domains or the impact of obvious data-quality errors.

Methods: We conducted a secondary cross-sectional analysis of a dataset available in the public domain claimed to consist of 541 women assessed in 10 hospitals in Kerala, India; 177 (32.7%) were classified as PCOS positive. Values that were highly implausible were retained as missing value using fairly conservative rules, but a sensitivity analysis was performed to exclude the complete nine records containing such values. Multivariable logistic regression was used to analyze clinical associations. Four ridge-logistic classification models were internally validated using repeated stratified 10-fold cross-validation (20 repeats): clinical (Model A), clinical plus biochemical (Model B), clinical plus biochemical plus ultrasound (Model C), and restricted clinical model not featuring prominent diagnostic-criterion features (Model D).

Results: In the adjusted clinical model (n = 538), OR 4.05, 95% CI 2.35-7.00 hair growth, OR 3.85, 95% CI 2.24-6.59 darkening of skin, aOR 3.67, 95% CI 2.08-6.45 irregularity of cycles, and OR 3.31, 95% CI 1.79-6.13 weight gain had strongest independent associations with PCOS status. Model A’s AUC was.881, 95% bootstrap CI was 0.848-0.912. Improvement in discrimination was not evident upon adding the biochemical variables (Model B AUC 0.879; 95% CI 0.845-0.909; delta AUC −0.002; 95% CI −0.008 to 0.004). By incorporating ultrasound variables, the Area Under Curve (AUC) improved to 0.949. Moreover, the most significant contributor to this gain was the isolated follicle count. The limited non-criteria clinical model still provided useful separation (AUC 0.843, 95% CI 0.805-0.877). When we omitted all nine records with grossly implausible values, results were nearly the same.

Conclusion: In this public dataset, readily available clinical features showed good discrimination, routine biochemical variables provided little incremental information, and ultrasound, in particular follicle count, greatly improved classification. As several influential variables coincide with the diagnostic criteria, the interpretation of these results is essentially, discrimination in a labelled dataset that already exists, not a prediction of a candidate or diagnostic validation which is independent.

Keywords: Polyendocrine Metabolic Ovarian Syndrome; Polycystic Ovary Syndrome; Ultrasound; Anti-Müllerian Hormone; Diagnostic Discrimination; Secondary Analysis

  1. Bozdag G., et al. “The prevalence and phenotypic features of polycystic ovary syndrome: A systematic review and meta-analysis”. Human Reproduction 12 (2016): 2841-2855.
  2. Lizneva D., et al. “Criteria, prevalence, and phenotypes of polycystic ovary syndrome”. Fertility and Sterility1 (2016): 6-15.
  3. Salari N., et al. “Global prevalence of polycystic ovary syndrome in women worldwide: A comprehensive systematic review and meta-analysis”. Archives of Gynecology and Obstetrics 3 (2024): 1303-1314.
  4. Lim SS., et al. “Overweight, obesity and central obesity in women with polycystic ovary syndrome: A systematic review and meta-analysis”. Human Reproduction Update6 (2012): 618-637.
  5. Lim SS., et al. “The effect of obesity on polycystic ovary syndrome: A systematic review and meta-analysis”. Obesity Reviews 2 (2013): 95-109.
  6. Kakoly NS., et al. “Ethnicity, obesity and the prevalence of impaired glucose tolerance and type 2 diabetes in PCOS: A systematic review and meta-regression”. Human Reproduction Update4 (2018): 455-467.
  7. Wekker V., et al. “Long-term cardiometabolic disease risk in women with PCOS: A systematic review and meta-analysis”. Human Reproduction Update6 (2020): 942-960.
  8. Teede HJ., et al. “Polyendocrine metabolic ovarian syndrome, the new name for polycystic ovary syndrome: A multistep global consensus process”. Lancet 10545 (2026): 2329-2339.
  9. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. “Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome”. Fertility and Sterility1 (2004): 19-25.
  10. Legro RS., et al. “Diagnosis and treatment of polycystic ovary syndrome: An Endocrine Society clinical practice guideline”. Journal of Clinical Endocrinology and Metabolism 12 (2013): 4565-4592.
  11. Teede HJ., et al. “Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome”. Journal of Clinical Endocrinology and Metabolism 10 (2023): 2447-2469.
  12. Lujan ME., et al. “Updated ultrasound criteria for polycystic ovary syndrome: Reliable thresholds for elevated follicle population and ovarian volume”. Human Reproduction 5 (2013): 1361-1368.
  13. Dewailly D., et al. “Definition and significance of polycystic ovarian morphology: A task force report from the androgen excess and polycystic ovary syndrome society”. Human Reproduction Update3 (2014): 334-352.
  14. Christ JP and Cedars MI. “Current guidelines for diagnosing PCOS”. Diagnostics6 (2023): 1113.
  15. Iliodromiti S., et al. “Can anti-Mullerian hormone predict the diagnosis of polycystic ovary syndrome? A systematic review and meta-analysis of extracted data”. Journal of Clinical Endocrinology and Metabolism 8 (2013): 3332-3340.
  16. Zigarelli A., et al. “Machine-aided self-diagnostic prediction models for polycystic ovary syndrome: Observational study”. JMIR Formative Research 3 (2022): e29967.
  17. Khanna VV., et al. “A distinctive explainable machine learning framework for detection of polycystic ovary syndrome”. Applied System Innovation 2 (2023): 32.
  18. Christodoulou E., et al. “A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models”. Journal of Clinical Epidemiology 110 (2019): 12-22.
  19. Collins GS., et al. “Transparent Reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): The TRIPOD statement”. Annals of Internal Medicine 1 (2015): 55-63.
  20. Collins GS., et al. “TRIPOD+AI statement: Updated guidance for reporting clinical prediction models that use regression or machine learning methods”. British Medical Journal 385 (2024): e078378.
  21. Wolff RF., et al. “PROBAST: A tool to assess the risk of bias and applicability of prediction model studies”. Annals of Internal Medicine 1 (2019): 51-58.
  22. Van Calster B., et al. “Calibration: The Achilles heel of predictive analytics”. BMC Medicine 17 (2019): 230.
  23. Steyerberg EW., et al. “Assessing the performance of prediction models: A framework for traditional and novel measures”. Epidemiology1 (2010): 128-138.
  24. Riley RD., et al. “Calculating the sample size required for developing a clinical prediction model”. British Medical Journal 368 (2020): m441.
  25. von Elm E., et al. “The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies”. Lancet 9596 (2007): 1453-1457.
  26. Azziz R., et al. “Polycystic ovary syndrome”. Nature Reviews Disease Primers 2 (2016): 16057.
  27. Escobar-Morreale HF. “Polycystic ovary syndrome: Definition, aetiology, diagnosis and treatment”. Nature Reviews Endocrinology 5 (2018): 270-284.
  28. Moran LJ., et al. “Lifestyle changes in women with polycystic ovary syndrome”. Cochrane Database of Systematic Reviews 7 (2011): CD007506.
  29. Cowan S., et al. “Lifestyle management in polycystic ovary syndrome - beyond diet and physical activity”. BMC Endocrine Disorders 23 (2023): 14.
  30. Kottarathil P. “Polycystic ovary syndrome (PCOS) [dataset on the Internet]”. Kaggle [date unknown].

Ahed J Alkhatib., et al. “Incremental Discrimination of Clinical, Biochemical, and Ultrasound Features for Polyendocrine Metabolic Ovarian Syndrome (Formerly PCOS): A Secondary Analysis of a Public Multicenter Dataset”. EC Gynaecology 15.10 (2026): 01-11.

ECronicon Journal

Our Services

Quick Links

Information

Creative Commons LicenseOpen Access by ECronicon is licensed under a Creative Commons Attribution 4.0 International License
Based on a work at www.ecronicon.net