PDF(Pubmed)
Abstract：
OBJECTIVE: To develop deep learning classifiers and evaluate their diagnostic performance in detecting the static gonioscopic angle closure and peripheral anterior synechia (PAS) based on swept source optical coherence tomography (SS-OCT) images.
METHODS: Subjects were recruited from the Glaucoma Service at Zhongshan Ophthalmic Center of Sun Yat-sun University, Guangzhou, China. Each subject underwent a complete ocular examination, such as gonioscopy and SS-OCT imaging. Two deep learning classifiers, using convolutional neural networks (CNNs), were developed to diagnose the static gonioscopic angle closure and to differentiate appositional from synechial angle closure based on SS-OCT images. Area under the receiver operating characteristic (ROC) 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, which contained 175 eyes of positive PAS, were recruited to develop diagnostic models. For the diagnosis of static gonioscopic angle closure, the first deep learning classifier achieved an AUC of 0.963 (95% CI, 0.954-0.972) with a sensitivity of 0.929 and a specificity of 0.877. The AUC of the second deep learning classifier distinguishing appositional from synechial angle closure was 0.873 (95% CI, 0.864-0.882) with a sensitivity of 0.846 and a specificity of 0.764.
CONCLUSIONS: Deep learning systems based on SS-OCT images showed good diagnostic performance for gonioscopic angle closure and moderate performance in the detection of PAS.
METHODS: Subjects were recruited from the Glaucoma Service at Zhongshan Ophthalmic Center of Sun Yat-sun University, Guangzhou, China. Each subject underwent a complete ocular examination, such as gonioscopy and SS-OCT imaging. Two deep learning classifiers, using convolutional neural networks (CNNs), were developed to diagnose the static gonioscopic angle closure and to differentiate appositional from synechial angle closure based on SS-OCT images. Area under the receiver operating characteristic (ROC) 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, which contained 175 eyes of positive PAS, were recruited to develop diagnostic models. For the diagnosis of static gonioscopic angle closure, the first deep learning classifier achieved an AUC of 0.963 (95% CI, 0.954-0.972) with a sensitivity of 0.929 and a specificity of 0.877. The AUC of the second deep learning classifier distinguishing appositional from synechial angle closure was 0.873 (95% CI, 0.864-0.882) with a sensitivity of 0.846 and a specificity of 0.764.
CONCLUSIONS: Deep learning systems based on SS-OCT images showed good diagnostic performance for gonioscopic angle closure and moderate performance in the detection of PAS.
摘要：
目的：开发深度学习分类器，并基于扫频源光学相干断层扫描（SS-OCT）图像评估其在检测静态前角镜闭角和外周前粘连（PAS）方面的诊断性能。
方法：从中山大学中山眼科中心青光眼中心招募受试者，广州,中国。每个受试者都接受了完整的眼部检查，如角度镜和SS-OCT成像。两个深度学习分类器，使用卷积神经网络(CNN)，已开发用于诊断静态前角镜角度闭合，并根据SS-OCT图像区分并置和粘连性角度闭合。使用受试者工作特征(ROC)曲线下面积(AUC)作为结果测量来评估两个深度学习系统的诊断性能。
结果：278例中国患者共439只眼，其中有175只眼睛的PAS阳性，被招募来开发诊断模型。对于静态房角镜闭角的诊断，第一个深度学习分类器的AUC为0.963（95％CI，0.954-0.972），灵敏度为0.929，特异性为0.877。第二个深度学习分类器区分并置与融合角闭合的AUC为0.873（95％CI，0.864-0.882），灵敏度为0.846，特异性为0.764。
结论：基于SS-OCT图像的深度学习系统显示出良好的房角镜闭角诊断性能和中等的PAS检测性能。
方法：从中山大学中山眼科中心青光眼中心招募受试者，广州,中国。每个受试者都接受了完整的眼部检查，如角度镜和SS-OCT成像。两个深度学习分类器，使用卷积神经网络(CNN)，已开发用于诊断静态前角镜角度闭合，并根据SS-OCT图像区分并置和粘连性角度闭合。使用受试者工作特征(ROC)曲线下面积(AUC)作为结果测量来评估两个深度学习系统的诊断性能。
结果：278例中国患者共439只眼，其中有175只眼睛的PAS阳性，被招募来开发诊断模型。对于静态房角镜闭角的诊断，第一个深度学习分类器的AUC为0.963（95％CI，0.954-0.972），灵敏度为0.929，特异性为0.877。第二个深度学习分类器区分并置与融合角闭合的AUC为0.873（95％CI，0.864-0.882），灵敏度为0.846，特异性为0.764。
结论：基于SS-OCT图像的深度学习系统显示出良好的房角镜闭角诊断性能和中等的PAS检测性能。
