PCOS Detection Using Machine Learning

Reproductive-age women worldwide suffer from metabolic problems, hormonal imbalance, and irregular menstruation due to PCOS. Complex symptoms of PCOS often lead to misdiagnosis or underdiagnosis, causing suffering and increasing the risk of obesity, diabetes, and cardiovascular disease. Treating these symptoms requires early, correct diagnosis. The project tests machine learning techniques such as Linear Regression, Ridge Regression, Lasso Regression, Random Forest Regressor, K-Nearest Neighbors, Decision Tree Regressor, and Support Vector Machines employing Mean Squared Error to address diagnostic issues. Since it handled noisy and non-linear data better, the Random Forest Regressor was best. Django-based internet applications and predictive algorithm help clinicians identify PCOS risk. The approach instantly assesses risk using BMI, age, blood pressure, and lifestyle factors. This simple method lets clinicians identify high-risk patients for rapid intervention and personalized treatment. Accuracy, scalability, and usability tests validated the system's clinical value. Finally, our machine learning-based solution will improve early PCOS identification, clinical resource use, and global women's health.