Data-driven biomarker discovery and risk profiling for polycystic ovary syndrome in Indian women using ensemble learning
Reproduction, Fertility and Development · 2025
DOI: 10.1071/rd25081 ↗Kurz gefasst
Fachartikel in der Zeitschrift Reproduction, Fertility and Development, 2025. Verfasst von Pulkit Verma und 3 weiteren Autor:innen. Beteiligte Einrichtung: Indian Council of Medical Research (ICMR) Headquarters, New Delhi, India. Email: pulkit.verma@icmr.gov.in. Hauptthemen laut MEDLINE: Polycystic Ovary Syndrome, Biomarkers, Machine Learning. Die Arbeit nennt 49 Literaturangaben.
Abstract (Originalsprache)
Context
Despite diagnostic advancements in India, the scarcity of Indian polycystic ovary syndrome (PCOS) data and varied diagnostic standards contribute to delays in PCOS detection, particularly in rural areas.
Aims
We aim to build a predictive model based on an extensive dataset derived from Indian studies and perform risk-based stratification of samples.
Methods
The PubMed database was queried for studies focused on the pathophysiology of PCOS in Indian women. Based on inclusion and exclusion criteria, six studies were selected. Corresponding clinical data was statistically synthesised based on study-specific baseline characteristics. The integrated dataset consisted of 11,258 samples (nPCOS = 7342; nControl = 3916) with 14 attributes: disease (PCOS vs control), age, body mass index, cholesterol, triglycerides, high-density and low-density lipoproteins, LH, FSH, testosterone, menarche age, systolic and diastolic blood pressure, and yearly menstrual cycles. After data pre-processing, missing values imputation, and feature engineering, model benchmarking was conducted using LazyPredict. LightGBM was selected for further hyperparameter tuning based on performance metrics. Lastly, feature importance analysis was performed, and predictive probabilities were utilised to categorise samples into different risk categories. Key results The optimised LightGBM model achieved 96.18% accuracy, 97.51% precision, 96.65% recall, and 99.31% receiver operating characteristic area under curve (ROC-AUC) score. Further, testosterone, menstrual cycles per year, triglycerides, LH, and diastolic blood pressure were the top five key attributes in PCOS. Risk categorisation of samples demonstrated substantial alignment with real diagnoses, validating the model’s clinical significance.
Conclusions
This study introduces the first comprehensively synthesised PCOS dataset for Indian women.
Implications
Our framework facilitates prompt risk detection, providing an adaptable methodology for decision-making in PCOS management.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Indian Council of Medical Research (ICMR) Headquarters, New Delhi, India. Email: pulkit.verma@icmr.gov.in.
- Amity Institute of Information Technology (AIIT), Amity University, Noida, Uttar Pradesh, India. Email: nsindhwani@amity.edu.
- ICMR - National Institute of Occupational Health (ICMR-NIOH), Gujarat, Ahmadabad, India. Email: sharma.lk@icmr.gov.in.
ORCID-Profile: Pulkit Verma, Rekha Agarwal
Zitieren
APA
Verma, P., Agarwal, R., Sharma, L. K., & Sindwani, N. (2025). Data-driven biomarker discovery and risk profiling for polycystic ovary syndrome in Indian women using ensemble learning. Reproduction, Fertility and Development. https://doi.org/10.1071/rd25081
Vancouver
Verma P, Agarwal R, Sharma LK, Sindwani N. Data-driven biomarker discovery and risk profiling for polycystic ovary syndrome in Indian women using ensemble learning. Reproduction, Fertility and Development. 2025. doi:10.1071/rd25081
BibTeX
@article{Verma2025,
author = {Verma, Pulkit and Agarwal, Rekha and Sharma, Lokesh Kumar and Sindwani, Nidhi},
title = {Data-driven biomarker discovery and risk profiling for polycystic ovary syndrome in Indian women using ensemble learning},
journal = {Reproduction, Fertility and Development},
year = {2025},
doi = {10.1071/rd25081}
}
EUVIMED