A Data-Driven Approach to PCOS Diagnosis: Systematic Review of Machine Learning Applications in Reproductive Health

AKSHAY V P; Dr. Ritvik Sriram; Dr. Keerthana R; Delna NS; Pranav Verma; Bhanu Verma; Mansi Trivedi; Shanmukhi Mogalipuvvu; Dr. Sasikala Kathiresan; Dr. Lalitha Soumya Johnson; Dr. Bhavit Bansal; Muhammed Asif; Shubhrit Shrivastava

doi:10.2478/amma-2025-0054

AKSHAY V P Department of Biotechnology, Manasarovar Global University, Bhopal, India
Dr. Ritvik Sriram Tirunelveli Medical College, Tirunelveli, Tamil Nadu, India
Dr. Keerthana R IQVIA Bengaluru, Karnataka, India
Delna NS BioDeskINDIA Labs, Bhopal, India
Pranav Verma Chitkara college of pharmacy, Chitkara University
Bhanu Verma University school of pharmaceutical sciences, Rayat Bahra University
Mansi Trivedi Humera Khan college of Pharmacy, University of Mumbai
Shanmukhi Mogalipuvvu Mamata academy of medical sciences, Hyderabad, Telangana, India
Dr. Sasikala Kathiresan Assistant professor, Department of Obstetrics and Gynaecology, All India Institute of Medical Science (AIIMS), Madurai, India
Dr. Lalitha Soumya Johnson Assistant professor, Department of Biotechnology and Microbiology, Dr. Janaki Ammal Campus, Kerala, India
Dr. Bhavit Bansal Senior Research Fellow, Central Council of Research in Yoga and Naturopathy, Delhi, Ministry of Ayush, India
Muhammed Asif Centennial College, Toronto, Canada
Shubhrit Shrivastava BioDeskINDIA Labs, Bhopal, India

DOI: https://doi.org/10.2478/amma-2025-0054

Keywords: Polycystic Ovary Syndrome, Machine Learning, Artificial Intelligence, PCOS Diagnosis

Abstract

Aim and Background: Polycystic Ovary Syndrome (PCOS) is a prevalent endocrine disorder in reproductive-aged women, characterized by hormonal imbalances, anovulation, and metabolic abnormalities. This systematic review aims to evaluate the effectiveness, types, and diagnostic performance of ML algorithms applied in PCOS detection and classification, and to identify the most frequently used input features and methodological challenges in existing studies.

Methods: A systematic search was conducted across PubMed, Scopus, and Google Scholar for studies published between 2014 and 2024 using keywords related to PCOS and machine learning. Inclusion criteria focused on original, peer-reviewed studies applying ML models for PCOS diagnosis. Data were extracted on model type, input features, diagnostic accuracy, and study design. Quality assessment was performed using the PROBAST tool.

Results: Out of 450 identified studies, 34 met the inclusion criteria and passed the quality assessment. Supervised learning models such as Random Forest, SVM, and XGBoost showed high accuracy (up to 99%). Deep learning approaches, particularly Convolutional Neural Networks (CNNs), achieved accuracies between 95% and 99.89% in analyzing ultrasound images. Hybrid models integrating clinical and imaging data further enhanced performance. Common input features included BMI, LH/FSH ratio, AMH, and ultrasound-based ovarian morphology. However, few studies validated models on external datasets, and input feature selection lacked standardization.

Conclusion: Machine learning models such as supervised, deep learning, and hybrid approaches show strong potential in improving PCOS diagnosis by identifying complex patterns across multi-dimensional datasets. Challenges such as limited generalizability and data standardization remain, therefore future studies should focus on developing explainable ML tools, validating models in clinical settings, and leveraging diverse data types for robust, personalized PCOS diagnosis.