Advancing PCOS Diagnosis: Harnessing the Power of AI and Machine Learning for Enhanced Accuracy and Efficiency

This research presents a diagnostic method for Polycystic Ovary Syndrome (PCOS), a common hormonal disorder among women of childbearing age. The study applied machine learning classifiers—Random Forest, CatBoost, and MLP—to the Kaggle PCOS dataset, enhanced by BorutaShap and SMOTE feature selection methods. The ensemble classifier achieved an F1 score of 93.54% and an accuracy of 96.71%. These results demonstrate AI's potential to improve PCOS diagnosis for broader clinical applications. Future research should integrate genetic and epigenetic factors into AI models, validated through clinical trials. Key contributions include: using an ensemble of machine learning classifiers, advanced feature selection methods, and achieving high F1 and accuracy scores, showing the model's clinical effectiveness.