AUTHOR=Yang Yangfan , Wu Yanyan , Guo Chong , Han Ying , Deng Mingjie , Lin Haotian , Yu Minbin TITLE=Diagnostic Performance of Deep Learning Classifiers in Measuring Peripheral Anterior Synechia Based on Swept Source Optical Coherence Tomography Images JOURNAL=Frontiers in Medicine VOLUME=Volume 8 - 2021 YEAR=2022 URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2021.775711 DOI=10.3389/fmed.2021.775711 ISSN=2296-858X ABSTRACT=Purpose: To develop deep learning classifiers and evaluate the diagnostic performance of which in detecting static gonioscopic angle closure and peripheral anterior synechia (PAS) based on swept source optical coherence tomography (SS-OCT) images. Method: Subjects were recruited from Glaucoma Service at Zhongshan Ophthalmic Center of Sun Yat-sun University, Guangzhou, China. Each subject underwent a complete ocular examination, including gonioscopy and SS-OCT imaging. Two deep learning classifiers, using convolutional neural network, were developed to differentiate static gonioscopic angle closure from open as well as appositional angle closure from synechial angle closure based on SS-OCT images. Area under the receiver operating characteristic curve (AUC) was used as outcome measure to evaluate the diagnostic performance of two deep learning systems. Results: A total of 439 eyes of 278 Chinese patients were recruited to develop diagnostic models. For differentiating static gonioscopic angle closure from angle open, the first classifier achieved an AUC of 0.963 (95% confidence interval, 0.954-0.972) with a sensitivity of 0.929 and a specificity of 0.877. While the AUC of the other classifier that distinguished appositional angle closure from synechial angle closure was 0.873 (95% confidence interval, 0.864-0.882) with a sensitivity of 0.846 and a specificity of 0.764. Conclusion: Deep learning systems based on SS-OCT images had efficient diagnostic performance in distinguishing gonioscopic angle closure from angle open but generally detected PAS.