Detection of Polycystic Ovary Syndrome

This paper introduces an automated deep learning algorithm for detecting polycystic ovary syndrome (PCOS) by analyzing scleral changes in eye images. The algorithm, applied to a dataset of 721 full-eye images from Chinese women, uses an improved U-Net for scleral image segmentation and a ResNet model for feature extraction. The multi-instance model achieved high performance metrics, with an average AUC of 0.979 and a classification accuracy of 0.929, demonstrating the effectiveness of deep learning in PCOS detection.