Deep Learning Algorithm for Automated Detection of Polycystic Ovary Syndrome Using Scleral Images.

Wenqi Lv,Ya Su,Ying Song,Xiangyu Jin,Qin Huang,Xue Lin,Xiaohui Shan,Lina Wang,Hao Zhong,Rongxin Fu,Wenli Du,Zhi Zhang,Guoliang Huang,Han Yang,Kai Jiang

doi:10.3389/fendo.2021.789878

Abstract

The high prevalence of polycystic ovary syndrome (PCOS) among reproductive-aged women has attracted more and more attention. As a common disorder that is likely to threaten women’s health physically and mentally, the detection of PCOS is a growing public health concern worldwide. In this paper, we proposed an automated deep learning algorithm for the auxiliary detection of PCOS, which explores the potential of scleral changes in PCOS detection. The algorithm was applied to the dataset that contains the full-eye images of 721 Chinese women, among which 388 are PCOS patients. Inputs of the proposed algorithm are scleral images segmented from full-eye images using an improved U-Net, and then a Resnet model was applied to extract deep features from scleral images. Finally, a multi-instance model was developed to achieve classification. Various performance indices such as AUC, classification accuracy, precision, recall, precision, and F1-score were adopted to assess the performance of our algorithm. Results show that our method achieves an average AUC of 0.979 and a classification accuracy of 0.929, which indicates the great potential of deep learning in the detection of PCOS.

Highlights

Polycystic ovary syndrome (PCOS) is known as one of the most common disorders among reproductive-aged women, affecting 6%–20% of premenopausal women worldwide [1]
To explore the feature extraction performance and the positive effects of attention mechanism applied in feature extraction network, we compared the detection performance of different Convolutional Neural Networks (CNNs) including Inception V3, Vgg16, and Vgg19 as feature extraction networks
As for PCOS detection, a feature extraction network was used for obtaining important deep features in the sclera, and later a multi-instance learning (MIL) model was applied to the final classification

Summary

Introduction

Polycystic ovary syndrome (PCOS) is known as one of the most common disorders among reproductive-aged women, affecting 6%–20% of premenopausal women worldwide [1]. The cardinal symptoms of PCOS are ovarian dysfunction and androgen excess. Factors such as genetics, puberty, physiological changes, mental state, and environmental influences are widely considered to induce this syndrome. Patients with PCOS frequently demonstrate menstrual irregularities, hirsutism, obesity, insulin resistance, and cardiovascular diseases [2]. Along with reproductive and metabolic disorders, a significant number of patients present psychological symptoms such as depression [3]. It is essential for the diagnosis and proper treatment of PCOS

Methods

Results

Discussion

Conclusion