Lifestyle and Symptom-Based Early Detection of Polycystic Ovary Syndrome Using Machine Learning

Polycystic Ovary Syndrome (PCOS) is a common endocrine disorder affecting women and often remains unnoticed in its early stages due to delayed medical consultation and dependence on laboratory-based diagnosis. This study presents an early detection approach for PCOS using machine learning based on lifestyle, menstrual, symptom-related, and other nonclinical parameters. A publicly available PCOS dataset from Kaggle containing 542 samples was used to identify early-stage PCOS risk. To support early detection, additional attributes such as symptom duration, stress level, and activity level were incorporated. The proposed approach analyzes indicators including body mass index, menstrual cycle patterns, lifestyle habits, and physical symptoms to assess PCOS risk at an early stage. Experimental results demonstrate that early detection of PCOS is feasible by effectively analyzing observable health symptoms and lifestyle factors.