A Data-Driven Approach to Polycystic Ovary Syndrome Diagnosis: Evaluating Machine Learning Models

This study demonstrates that machine learning can significantly enhance the early and accurate detection of Polycystic Ovary Syndrome (PCOS) by analyzing clinical and demographic data. Using a dataset of 539 women, including 176 PCOS-positive cases, the study evaluated six machine learning models, with Random Forest achieving the highest performance with 93% accuracy and 86% sensitivity. Key predictive features included antral follicle count, hair growth, skin pigmentation, weight gain, and fast-food consumption. The findings suggest that integrating machine learning models into clinical practice could offer a cost-effective and efficient alternative to traditional diagnostic methods, facilitating timely interventions and improving patient outcomes.