Center for Clinical Research and Biostatistics, Jockey Club School of Public Health and Primary Care, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong, China.
Department of Ophthalmology, Zhongshan Hospital, Fudan University, Shanghai, China.
BMJ Open Ophthalmol. 2024 Sep 10;9(1):e001594. doi: 10.1136/bmjophth-2023-001594.
Train an automatic retinal image analysis (ARIA) method to screen glaucomatous optic neuropathy (GON) on non-mydriatic retinal images labelled with the additional results of optical coherence tomography (OCT) and assess different models for the GON classification.
All the images were obtained from the hospital for training and 10-fold cross-validation. Two methods were used to improve the classification performance: (1) using images labelled with the additional results of OCT as the reference standard and (2) generating models using retinal features from the entire images, the region of interest (ROI) of the optic disc, and the ROI of the macula, and the combination of all the features.
Overall, we collected 1338 images with paired OCT scans. In 10-fold validation, ARIA achieved sensitivities of 92.2 %, 92.7% and 85.7%, specificities of 88.8%, 86.7% and 80.2% and accuracies of 90.6%, 89.9% and 83.1% using the retinal features from the entire images, the ROI of the optic disc and the ROI of the macula, respectively. We found the model combining all the features has the best classification performance and obtained a sensitivity of 92.5%, a specificity of 92.1% and an accuracy of 92.4%, which is significantly different from other models (p<0.001).
We used two methods to improve the classification performance and found the best model to detect glaucoma on colour fundus retinal images. It can become a cost-effective and relatively more accurate glaucoma screening tool than conventional methods.
训练一种自动视网膜图像分析（ARIA）方法，以对未经散瞳的视网膜图像进行筛查，这些图像已用光学相干断层扫描（OCT）的附加结果进行标记，并评估用于青光眼视神经病变（GON）分类的不同模型。
所有图像均来自医院进行训练和 10 倍交叉验证。使用两种方法来提高分类性能：（1）使用用 OCT 的附加结果标记的图像作为参考标准；（2）使用整个图像、视盘感兴趣区域（ROI）和黄斑 ROI 的视网膜特征生成模型，以及所有特征的组合。
总体而言，我们共收集了 1338 对带有 OCT 扫描的图像。在 10 倍验证中，ARIA 使用整个图像、视盘 ROI 和黄斑 ROI 的视网膜特征分别实现了 92.2%、92.7%和 85.7%的敏感性、88.8%、86.7%和 80.2%的特异性以及 90.6%、89.9%和 83.1%的准确性。我们发现结合所有特征的模型具有最佳的分类性能，获得了 92.5%的敏感性、92.1%的特异性和 92.4%的准确性，与其他模型有显著差异（p<0.001）。
我们使用两种方法来提高分类性能，并找到了最佳模型，以在彩色眼底视网膜图像上检测青光眼。它可以成为比传统方法更具成本效益和相对更准确的青光眼筛查工具。
