Background: The supervised deep learning approach provides state-of-the-art performance in a variety of fundus image classification tasks, but it is not applicable for screening tasks with numerous or unknown disease types. The unsupervised anomaly detection (AD) approach, which needs only normal samples to develop a model, may be a workable and cost-saving method of screening for ocular diseases. Objective: This study aimed to develop and evaluate an AD model for detecting ocular diseases on the basis of color fundus images. Methods: A generative adversarial network–based AD method for detecting possible ocular diseases was developed and evaluated using 90,499 retinal fundus images derived from 4 large-scale real-world data sets. Four other independent external test sets were used for external testing and further analysis of the model’s performance in detecting 6 common ocular diseases (diabetic retinopathy [DR], glaucoma, cataract, age-related macular degeneration, hypertensive retinopathy [HR], and myopia), DR of different severity levels, and 36 categories of abnormal fundus images. The area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the model’s performance were calculated and presented. Results: Our model achieved an AUC of 0.896 with 82.69% sensitivity and 82.63% specificity in detecting abnormal fundus images in the internal test set, and it achieved an AUC of 0.900 with 83.25% sensitivity and 85.19% specificity in 1 external proprietary data set. In the detection of 6 common ocular diseases, the AUCs for DR, glaucoma, cataract, AMD, HR, and myopia were 0.891, 0.916, 0.912, 0.867, 0.895, and 0.961, respectively. Moreover, the AD model had an AUC of 0.868 for detecting any DR, 0.908 for detecting referable DR, and 0.926 for detecting vision-threatening DR. Conclusions: The AD approach achieved high sensitivity and specificity in detecting ocular diseases on the basis of fundus images, which implies that this model might be an efficient and economical tool for optimizing current clinical pathways for ophthalmologists. Future studies are required to evaluate the practical applicability of the AD approach in ocular disease screening.

This is the abstract only. Read the full article on the JMIR site. JMIR is the leading open access journal for eHealth and healthcare in the Internet age.

中文翻译：

---

异常检测模型在使用彩色视网膜眼底图像筛查眼病中的应用：设计和评估研究

背景：监督式深度学习方法在各种眼底图像分类任务中提供了最先进的性能，但不适用于具有多种或未知疾病类型的筛选任务。仅需要正常样本来开发模型的无监督异常检测 (AD) 方法可能是一种可行且节省成本的眼部疾病筛查方法。目的：本研究旨在开发和评估基于彩色眼底图像检测眼部疾病的 AD 模型。方法：使用来自 4 个大规模真实世界数据集的 90,499 个视网膜眼底图像开发和评估了一种基于生成对抗网络的 AD 方法，用于检测可能的眼部疾病。使用另外四个独立的外部测试集进行外部测试并进一步分析模型在检测 6 种常见眼部疾病（糖尿病视网膜病变 [DR]、青光眼、白内障、年龄相关性黄斑变性、高血压性视网膜病变 [HR] 和近视）方面的性能，不同严重程度的DR，以及36类异常眼底图像。计算并呈现模型性能的受试者工作特征曲线下面积 (AUC)、准确性、灵敏度和特异性。结果：我们的模型在内部测试集中检测异常眼底图像的 AUC 为 0.896，灵敏度为 82.69%，特异性为 82.63%，在 1 个外部专有数据集中，AUC 为 0.900，灵敏度为 83.25%，特异性为 85.19%。在检测 6 种常见眼科疾病中，DR 的 AUC，青光眼、白内障、AMD、HR 和近视分别为 0.891、0.916、0.912、0.867、0.895 和 0.961。此外，AD 模型检测任何 DR 的 AUC 为 0.868，检测可参考 DR 的 AUC 为 0.908，检测威胁视力的 DR 的 AUC 为 0.926。结论：AD方法在基于眼底图像检测眼部疾病方面实现了高灵敏度和特异性，这意味着该模型可能是优化眼科医生当前临床路径的有效且经济的工具。未来的研究需要评估 AD 方法在眼部疾病筛查中的实际适用性。926 用于检测威胁视力的 DR。结论：AD方法在基于眼底图像检测眼部疾病方面实现了高灵敏度和特异性，这意味着该模型可能是优化眼科医生当前临床路径的有效且经济的工具。未来的研究需要评估 AD 方法在眼部疾病筛查中的实际适用性。926 用于检测威胁视力的 DR。结论：AD方法在基于眼底图像检测眼部疾病方面实现了高灵敏度和特异性，这意味着该模型可能是优化眼科医生当前临床路径的有效且经济的工具。未来的研究需要评估 AD 方法在眼部疾病筛查中的实际适用性。

这只是摘要。阅读 JMIR 网站上的完整文章。JMIR 是互联网时代电子健康和医疗保健领域领先的开放获取期刊。