| Abstrakt: |
Polycystic ovary syndrome (PCOS) is an endocrine disorder characterized by excess androgen hormone and symptoms like obesity, irregular cycles, acne, hirsutism, and abnormal ovarian follicle development. Treatment targets anovulation, hyper androgenism, and menstrual dysfunction. This study developed machine learning models using clinical data to predict PCOS presence. The dataset included FSH/LH values and observed symptoms. Random Forest, Decision, SVM, Naïve Bayesian, Gradient Boosting, and KNN classifiers were evaluated. Random Forest achieved 90% accuracy. Additionally, a doctor model based on CNN deep learning will provide a secondary diagnostic opinion to professionals. This model will enhance the accuracy of PCOS diagnosis and support healthcare providers in decision-making. An assistive platform is proposed to detect PCOS, provide prevention measures, and offer education on menstrual health and well-being. It aims to benefit rural communities and address specific PCOS needs in a concise manner. [ABSTRACT FROM AUTHOR] |
