Prediction of Ovarian Syndrome Using Deep Feature Representation and Learning Approaches

The high incidence of the polycystic ovarian syndrome (PCOS) among fertile women receives more and more attention. Women's physical and mental health may be at risk due to the common disorder PCOS, whose identification is an expanding issue in worldwide public health. In this work, we looked into the potential significance of abnormalities in the identification of PCOS and provided information-based feature representation and learning techniques (IFRL) for PCOS auxiliary identification. Here, a real-time dataset is acquired from the Coimbatore region by examining 74 individuals, and these samples are used in the algorithm. The suggested method performs pre-processing, feature learning and classification using the proposed IFRL. To achieve categorization, a multi-instance model was finally built. Several performance indices were used to evaluate our method's effectiveness, including AUC, precision, recall, accuracy, and F1-score for classification.