Developing A System for Automatic Prediction of Polycystic Ovary Syndrome Using Machine Learning

Abstract

Polycystic Ovary Syndrome (PCOS) is a condition that leads to lifelong health problems outside of infertility. The lack of a single, known cause and universal symptoms makes diagnosis challenging. The early and accurate prediction will prevent many subsequ ent serious and morbid illnesses that can arise from PCOS. Therefore, this study proposes a predictive Machine Learning (ML) model to identify patients at risk of PCOS and alert healthcare professionals, allowing for early intervention. The predictive performance of the Random Forest Classifier, Logistic Regression, Gradient Boost, Adaptive Boost, and XGBoost machine learning algorithms was compared based on accuracy, precision, recall, and F1-score with a publicly available dataset from Kaggle. The experiment was performed in Google Colab. Our experimental results showed a good predictive performance of 90% across all evaluation metrics. However, Random Forest outperformed all other models achieving 96% accuracy, precision, recall, and F1-score, respectively. The high accuracy (96%) obtained by this study suggests that the proposed model could effectively identify patients at risk of PCOS, potentially aiding early diagnosis and intervention. CCS CONCEPTS • Applied computing • Life and Medical sciences • Health Informatics